RNF217: brokering ferroportin degradation

In this issue of Blood, Jiang and colleagues identify an E3 ubiquitin ligase that promotes the degradation of the iron efflux protein ferroportin (FPN).¹  FPN is the only known cellular iron exporter; as such, it plays a central role in maintaining intracellular iron homeostasis as well as regulating systemic iron metabolism.²  FPN regulation is particularly important in modulating dietary iron absorption (duodenal enterocytes) and iron redistribution (from reticuloendothelial macrophages, hepatocytes, erythrocytes). The master iron regulatory hormone hepcidin functions by binding to a specific FPN extracellular loop, preventing iron efflux, and promoting FPN internalization and degradation.³  The hepcidin-induced internalization of FPN is dependent on ubiquitination of certain lysine residues on the third intracellular loop.^4,5  Early studies attempting to characterize the molecular players in FPN degradation led to erroneous conclusions regarding the participation of tyrosine kinases and certain E3 ligases. Despite nearly 10 years since the discovery that hepcidin binding leads to FPN ubiquitination, the identity of the responsible ubiquitin ligase(s) has remained a mystery. The studies by Jiang and colleagues support the participation of ring finger protein 217 (RNF217).

The investigators’ route to RNF217 was indirect. They initially focused on the epigenetic changes in murine liver samples in response to manipulations of dietary iron intake. During these studies, an RNA screen identified upregulation of the DNA demethylase Tet1 in liver of mice fed a high-iron diet. They generated mice with global Tet1 knockout and observed consequences on iron metabolism when placed on a high-iron diet. Compared with wild-type mice on the same diet, the Tet1^−/− mice demonstrated increased serum and liver iron, splenic iron sparing, and greater duodenal and splenic macrophage expression of FPN protein. In these regards, the mice manifested the classic hereditary hemochromatosis (HH) phenotype caused by low hepcidin; however, liver hepcidin expression in the Tet1^−/− mice was instead found to be higher. These findings are similar to those observed in a rare form of HH caused by FPN gain-of-function mutations that result in resistance to hepcidin-mediated degradation.⁶  Among other findings consistent with excess FPN, the investigators found that FPN ubiquitination was diminished in response to hepcidin in bone marrow–derived macrophages (BMDMs) from Tet1^−/− mice. They then screened for candidate E3 ubiquitin ligases that (1) interacted with FPN, (2) were differentially expressed by Tet1 knockdown, and (3) were downregulated in macrophages from Tet1^−/− mice. This strategy identified RNF217, which was then shown to be hypermethylated in the Tet1^−/− mice.

RNF217 has not been investigated much. The RNF217 gene (also known as OSTL, opposite STL) encodes a 284-amino-acid protein with 2 RING finger domains.⁷  Many proteins containing RING finger domains function as E3 ubiquitin ligases, although this property had not previously been studied for RNF217. RNF217 is expressed in multiple tissues, has been reported to interact with the antiapoptotic protein HAX1, and is overexpressed in myeloid leukemia cells.⁷  Global Rnf217 knockout mice have been generated, the characterization of which is limited to that reported by the vendor. However, no significant phenotype is reported.⁸ 

Jiang et al found that RNF217 served to polyubiquitinate FPN when each were overexpressed in HEK293 cells. Moreover, RNF217-mediated FPN degradation was abrogated by proteasomal and lysosomal inhibition. However, FPN polyubiquitination was increased by RNF217 overexpression regardless of hepcidin treatment; conversely, polyubiquitination of FPN was increased by hepcidin treatment regardless of RNF217 overexpression. These observations raise concerns for epiphenomena and indicate the need for additional studies at endogenous expression levels in relevant cell types. The investigators thus generated mice with cell-specific Rnf217 knockout, directing Cre recombinase with the LysM promoter (which targets tissue macrophages and a proportion of splenic macrophages) and the Villin promoter (which targets enterocytes). No significant phenotype was observed at 2 months of age in either model. However, by 5 months, the Rnf217^Lysm/Lysm mice demonstrated decreased splenic iron. Importantly, FPN stability was greater in BMDMs from Rnf217^Lysm/Lysm compared with wild-type mice upon ex vivo treatment with hepcidin. This relative hepcidin resistance was also demonstrated in vivo. Exogenous hepcidin mediated a decrease in splenic FPN and fall in serum iron in wild-type but not Rnf217^Lysm/Lysm mice. The hypoferremic response was likewise attenuated in Rnf217^Lysm/Lysm mice in response to endotoxin-mediated upregulation of endogenous hepcidin. The enterocyte knockout (Rnf217^{villii/villin}) mice at 6 months demonstrated increased duodenal FPN, decreased duodenal iron, and increased hepatic iron, consistent with attenuated intestinal FPN degradation and excess enterocyte iron transfer.

The magnitude of iron loading in each of the tested models is much less than reported in mice in which the hepcidin-binding site of Fpn had been mutated (Slc40a1^C326S/C326S).⁹  Possibly mice with combined or complete knockout of Rnf217 would manifest a more similar phenotype. Perhaps ubiquitination accounts for only a portion of hepcidin-mediated FPN downregulation, with the remainder consequent to hepcidin binding–dependent inactivation of iron export function. Such FPN inactivation would be abrogated with loss of hepcidin binding, but intact despite loss of Rfn217. It is also possible that RNF217 is not the only E-3 ubiquitin ligase participating in FPN degradation. Indeed, other candidates were identified by Jiang et al in their screenings. It is important to recognize that FPN is regulated at multiple levels²  (see table) and compensatory mechanisms appear be operative in the studies reported here, as evidenced by decreased Slc40a1 messenger RNA (encoding FPN) in BMDMs from Tet1^−/− and Rnf217^Lysm/Lysm mice. Future studies integrating these considerations, along with the indexing of splenic and enterocytic FPN levels to hepcidin levels, will be useful in further defining the contribution of RNF217 to FPN regulation.

Dysregulation of the hepcidin-FPN axis underlies several clinical conditions. There has been speculation that identification of the E3 ligase(s) mediating FPN degradation might provide a novel therapeutic target.⁵  The broad tissue distribution of RNF217 and its potential role in regulating apoptosis raise concern for off-target effects. Regardless, the studies of Jiang et al identify a role for RNF217 in FPN ubiquitination and stability. They moreover demonstrate a role for the iron-responsive demethylase Tet1 in iron metabolism that plausibly includes the regulation of Rnf217. Their findings provide important insights, and a foundation for continued investigation, into hepcidin-mediated degradation of FPN.