What does the FOX(O) say? High iron: hepcidin!

In this issue of Blood, Xu et al have identified FOXO1, a transcription factor commonly associated with glucose metabolism, as an important player in systemic iron regulation by its activation of hepcidin transcription in the liver.¹ This study shows that FOXO1 regulates the hepatocyte’s response to high iron conditions or iron overload. It bridges a knowledge gap in how fluctuations in serum iron levels regulate the expression of hepcidin, a key iron-regulatory hormone that controls iron export from the gut and storage tissues.² 

The link between iron sensing and control of hepcidin expression is complex. When serum iron levels are high, the liver increases hepcidin production. In turn, hepcidin either occludes the iron exporter, ferroportin, or induces its ubiquitination and internalization (reviewed by Galy et al²). The hepcidin-ferroportin axis thus decreases iron export from the gut and storage tissues into circulation when iron levels are high. Conversely, downregulation of hepcidin signaling increases iron availability when serum iron is low. Hepcidin expression is activated by BMP signaling and SMAD 1/5/8 phosphorylation, which then directly upregulates hepcidin transcription.³ Hepcidin transcription is also upregulated via the interaction of HFE (homeostatic iron protein) with the BMP/SMAD pathway. This signaling complex is modulated by TFR1 binding, providing a mechanism connecting iron availability to BMP/SMAD signaling and downstream control of hepcidin expression.^4,5 However, as BMP/SMAD signaling is involved in a plethora of biological processes (in addition to iron regulation), are there specific mechanisms that integrate BMP signaling with iron levels to control hepcidin transcription?

This work showed that elevated iron drives nuclear localization of FOXO1, which is a key mechanism governing FOXO1 transcriptional activity. Using pharmacological methods and tissue-specific or -inducible loss of function models, the authors found that loss of hepatic FOXO1 decreased BMP-dependent SMAD 1/5/8 phosphorylation, decreasing hepcidin transcription. Consequently, animals deficient in hepatic FOXO1 maintained elevated ferroportin expression in the small intestine, resulting in increased tissue iron stores. FOXO1 was also required for regulation of hepcidin expression via HFE, and FOXO1 overexpression alleviates iron overload in a Hfe^−/− model of hereditary hemochromatosis, likely via its ability to increase SMAD phosphorylation and nuclear localization, leading to downstream hepcidin activation (see figure). Collectively, these data suggest that small molecules that modulate FOXO1 activity have potential as a therapeutic strategy for disorders related to iron availability, particularly iron overload.

Immunoprecipitation experiments suggest a mechanism in which hepatocyte FOXO1 facilitates optimal phosphorylation and activation of SMADs and subsequently interacts with pSMADs. The FOXO1-pSMAD complex is predicted to translocate to the nucleus, where the 2 proteins bind in proximity to DNA regions near the hepcidin start site to activates hepcidin transcription. Although this work does not yet provide direct mechanistic insights into how FOXO1 interacts with the BMP/SMAD pathway to specifically activate hepatic hepcidin transcription in response to increased serum iron, immunoprecipitation experiments indicate FOXO1 is required for optimal phosphorylation of cytosolic SMAD. Several potential mechanisms include possible stimulation of BMP receptor kinase activity by cytosolic FOXO1 or that the FOXO1/SMAD interaction facilitates SMAD phosphorylation via protein conformational changes. Follow-up studies that interrogate the details of the FOXO1/SMAD interaction, FOXO1’s role in regulation of SMAD phosphorylation, and the role of serum or intracellular iron in this pathway will be important for dissecting the role of the FOXO/SMAD signaling node in cytosolic iron sensing, as well as selective activation of BMP/SMAD targets in response to specific stimuli. Of relevance to signaling specificity, regulation of FOXO1 posttranslational modifications appears to be tissue specific,⁶ providing additional layers of nuance in the specificity of the FOXO1/SMAD signaling node.

The discovery of hepatic FOXO1 as an iron regulator is exciting as it breaks mechanistic ground in understanding how iron and glucose metabolism are integrated to optimize energy availability and utilization.^6,7 Xu et al conducted careful studies that revealed sex-specific aspects of FOXO1/SMAD signaling. Although the effects of FOXO1 on hepcidin expression tended to be most pronounced in animals on high iron diets, hepatic hepcidin mRNA expression in female animals on normal diets were more sensitive to FOXO1 perturbations than in male animals. However, serum hepcidin in male animals was decreased in FOXO1 deficiency, but this was not observed in female animals. These data will be important to consider when carrying out follow-up studies, particularly with relevance to understanding iron regulation and erythropoiesis during energy-intensive, sex-specific processes such as pregnancy, birth, and lactation, which dramatically increase iron and erythropoietic demand, requiring the use of additional iron regulatory mechanisms.⁸ The relevance of FOXO1 in pathological iron overload contexts such as sickle cell disease and porphyria will be also be important question to address, given its importance in hepcidin activation during iron overload.

These rigorous studies provide important contextual information on regulation of systemic iron levels. Careful biochemical studies in tissue-specific contexts and disaggregation of data from male and female animals provide convincing evidence that FOXO1 (and potentially other FOXO transcription factors) play key roles in regulating iron metabolism within specific physiologic contexts. Transcriptomic and chromatin immunoprecipitation sequencing studies of FOXO1 and SMAD transcription targets will likely be informative regarding the role of FOXO1 and SMAD integrating iron metabolism with other nutrient metabolism pathways. Critically, the iron studies in this article focused on nonheme iron and did not consider bioavailable forms of iron such as heme and iron-sulfur clusters. The next level of mechanistic insight integrating biochemical pathways via nodes such as FOXO1 will be gleaned from careful interrogation of iron fate in response to FOXO1 perturbation and concomitant alterations in other metabolic pathways.

Conflict-of-interest disclosure: Y.Y.Y. declares no competing financial interests.