Induced Disruption of the Iron-Regulatory Hormone Hepcidin Inhibits Acute Inflammatory Hypoferraemia

Abstract

Withdrawal of iron from serum (hypoferraemia) is a conserved innate immune antimicrobial strategy that can withhold this critical nutrient from invading pathogens, impairing their growth. Hepcidin (Hamp1) is the master regulator of iron and its expression is induced by inflammation. Mice lacking Hamp1 from birth rapidly accumulate iron and are susceptible to infection by blood-dwelling siderophilic bacteria such as Vibrio vulnificus. In order to study the innate immune role of hepcidin against a background of normal iron status, we developed a transgenic mouse model of tamoxifen-sensitive conditional Hamp1 deletion (termed iHamp1-KO mice). These mice attain adulthood with an iron status indistinguishable from littermate controls. Hamp1 disruption and the consequent decline of serum hepcidin concentrations occurred within hours of a single tamoxifen dose. We found that the TLR ligands LPS and Pam3CSK4 and heat-killed Brucella abortus caused an equivalent induction of inflammation in control and iHamp1-KO mice. Pam3CSK4 and B. abortus only caused a drop in serum iron in control mice, while hypoferraemia due to LPS was evident but substantially blunted in iHamp1-KO mice. Our results characterise a powerful new model of rapidly inducible hepcidin disruption, and demonstrate the critical contribution of hepcidin to the hypoferraemia of inflammation.

Fig. 1

Induction of targeted disruption of Hamp1 following tamoxifen administration to iHamp1-KO mice. Seven-week-old male iHamp1-KO mice and littermate iHamp1-Ctrl mice were treated daily for 5 days with 1 mg of tamoxifen per mouse; 3 months later, the mice were sacrificed and tissues taken for analysis. a Genotyping PCR from ear tissue confirming the presence of the Cre allele (upper panel) in iHamp1-KO but not iHamp1-Ctrl mice (lower panel) despite the presence of the flox-Hamp1 allele in both (761 bp). b Genotyping PCR confirming deletion of Hamp1 exons 2-3 in the ear, liver, duodenum and spleen in tamoxifen-induced iHamp1-KO mice, but not iHamp1-Ctrl mice. c Significant depletion of Hamp1 mRNA in the liver, lung, spleen and duodenum in tamoxifen-induced iHamp1-KO mice compared to iHamp1-Ctrl mice, quantified by qRT-PCR (dot plots show the geometric mean ± geometric SD, with p values reflecting the results of t tests based on log-transformed data). d Significant decrease in serum hepcidin in tamoxifen-induced iHamp1-KO mice. Dot plot shows the arithmetic mean ± SD; t test.

Fig. 2

Evidence of iron loading 3 months after tamoxifen-administration to iHamp1-KO mice. Seven-week-old male iHamp1-KO mice and littermate iHamp1-Ctrl mice were treated daily for 5 days with 1 mg of tamoxifen per mouse; 3 months later, the mice were sacrificed and tissues taken for analysis. Comparisons between tamoxifen-induced iHamp1-KO and littermate iHamp1-Ctrl mice. a Perls' staining of 4-µm liver sections indicating no iron loading around the terminal venule (TV) in iHamp1-Ctrl mice (left), but increasing iron loading towards the TV compared to the portal triad (PT) in induced iHamp1-KO mice (right). b Non-heme liver iron quantification. c Perls' staining of 4-µm spleen sections with macrophage iron staining in iHamp1-Ctrl spleens (left) indicated with yellow arrows, absent in induced iHamp1-KO mice. d Non-heme spleen iron quantification. e Elevated Bmp6 mRNA. f Serum iron. g Tsat (the dataset includes imputed Tsat values: when UIBC was below limit of detection, Tsat was allocated the value of 100%) h Serum ferritin in iHamp1-KO mice. Images were obtained using a Nikon DS-FI1 camera with a Nikon DS-L2 control unit and an Olympus BX40 microscope. Dot plots show the arithmetic mean ± SD; p values indicate the results of t tests. For colors, see online version.

Fig. 3

Effect of the tamoxifen dose and schedule on Hamp1 disruption in iHamp1-KO mice. a-d Seven- to eight-week-old male iHamp1-KO mice were treated with 1 dose of 1 mg of tamoxifen per day, daily for between 1 and 4 days (i.e. 1 dose per day for up to 4 days); littermate iHamp1-Ctrl mice were given 1 tamoxifen dose each day for 4 consecutive days. The mice were sacrificed and tissues taken for analysis 3 days after the last tamoxifen dose. Changes in liver Hamp1 mRNA (p < 0.0001; a), serum hepcidin (p = 0.0002; b), serum iron (p < 0.0001; c), and Tsat (p = 0.015; d) with increasing numbers of tamoxifen doses. Dot plots show the arithmetic mean ± SD (except a, which shows the geometric mean ± geometric SD, with statistics based on log-transformed data); p values represent the results of one-way ANOVA, with Dunnett's multiple comparison test (*** p < 0.001, ** p < 0.01, * p < 0.05; ^# dataset includes imputed Tsat values: when UIBC was below the limit of detection, Tsat was allocated the value of 100%). e-h Seven- to ten-week-old male iHamp1-KO mice and littermate iHamp1-Ctrl mice were administered a single 1 mg/mouse tamoxifen dose; mice were sacrificed and tissues taken for analysis after 6, 24 or 72 h. e PCR detection of Hamp1 deletion in tamoxifen-induced iHamp1-KO mice, indicated by the presence of a Cre allele. f Reduction in liver Hamp1 mRNA in iHamp1-KO mice as early as 6 h post-tamoxifen administration. Changes in serum hepcidin (g), and serum iron (h). Dot plots show the mean ± SD (except e, which shows the geometric mean ± geometric SD, with statistics based on log-transformed data); p values indicate the results of t tests.

Fig. 4

Effect of tamoxifen-induced hepcidin disruption on the response to LPS in iHamp1-KO mice. Seven- to ten-week-old male iHamp1-KO mice and littermate iHamp1-Ctrl mice were administered 1 mg of tamoxifen per mouse 24 h prior to being given 1 µg/g of LPS (E. coli O55:B5) or PBS vehicle intraperitoneally; mice were culled 6 h after LPS administration. Induction of liver Il6 (a) and Fga (b) mRNA, and reduction of Slc40a1 mRNA (c) in both tamoxifen-treated iHamp1-Ctrl and iHamp1-KO mice treated with LPS. Lack of Hamp1 mRNA (d) and serum hepcidin (e) response to LPS in tamoxifen-treated iHamp1-KO mice in contrast to tamoxifen-treated iHamp1-Ctrl mice. f Lack of absolute hypoferraemic response to LPS in tamoxifen-treated iHamp1-KO mice. Dot plots show the mean ± SD; p values indicate the results of t tests; in cases where data are spread over orders of magnitude, data are plotted with log scales, geometric means ± geometric SD are shown, and t tests are performed on log-transformed data.

Fig. 5

Lack of hypoferraemic response to the TLR1/2 agonist Pam3CSK4 in tamoxifen-induced iHamp1-KO mice. Eight- to nine-week-old male iHamp1-KO mice and littermate iHamp1-Ctrl mice were administered 1 mg of tamoxifen per mouse 24 h prior to being given 400 ng/g Pam3CSK4 or PBS vehicle intraperitoneally; mice were culled 3 h after LPS administration. Induction of liver Il6 (a) and Fga (b) mRNA, and reduction of Slc40a1 mRNA (c) in both tamoxifen-treated iHamp1-Ctrl and iHamp1-KO mice treated with Pam3CSK4. Lack of Hamp1 mRNA (d) and serum hepcidin (e) response to Pam3CSK4 in tamoxifen-treated iHamp1-KO mice in contrast to tamoxifen-treated iHamp1-Ctrl mice. f Lack of hypoferraemic response to Pam3CSK4 in tamoxifen-treated iHamp1-KO mice. Dot plots show the mean ± SD; p values indicate the results of t tests; in cases where data are spread over orders of magnitude, data are plotted with log scales, geometric means ± geometric SD are shown, and t tests are performed on log-transformed data.

Fig. 6

No hypoferraemic response to HKBA in tamoxifen-induced iHamp1-KO mice. Eight- to nine-week-old female iHamp1-KO mice and littermate iHamp1-Ctrl mice were administered 1 mg of tamoxifen per mouse 24 h prior to being given 2.5 × 10⁸ HKBA or PBS vehicle intraperitoneally; mice were culled 6 h after HKBA administration. Induction of liver Il6 (a) and Fga (b) mRNA, and reduction of Slc40a1 mRNA (c) in both tamoxifen-treated iHamp1-Ctrl and iHamp1-KO mice treated with HKBA. Lack of Hamp1 (d) mRNA and serum hepcidin (e) response to HKBA in tamoxifen-treated iHamp1-KO mice in contrast to tamoxifen-treated iHamp1-Ctrl mice. f Lack of hypoferraemic response to HKBA in tamoxifen-treated iHamp1-KO mice. Dot plots show the mean ± SD; p values indicate the results of t tests; in cases where data are spread over orders of magnitude, data are plotted with log scales, geometric means ± geometric SD are shown, and t tests are performed on log-transformed data.