Minihepcidin peptides as disease modifiers in mice affected by β-thalassemia and polycythemia vera

In β-thalassemia, mutations in the β-globin gene lead to a relative excess of α-globin, which, together with heme, forms hemichromes, molecules with toxic potential to trigger reactive oxygen species (ROS), apoptosis of erythroid progenitors, and short-lived erythrocytes.^1-3  This results in anemia, chronically high levels of erythropoietin, and ineffective erythropoiesis—overproduction of erythroid progenitors that fail to generate mature erythrocytes.⁴  Even when transfusions are not required, the manifestations of the disease can be severe, including chronic anemia and splenomegaly as well as pulmonary hypertension and an increased risk of thrombosis.⁴  Anemia and ineffective erythropoiesis also cause increased iron absorption mediated by duodenal hypoxia and erythroid-mediated suppression of the iron regulator hepcidin.^5-10  The resulting iron overload adds to morbidity by causing cirrhosis, risk of cardiomyopathy, and hepatocarcinoma and exacerbating erythroid cell damage, apoptosis, and ineffective erythropoiesis.¹  Even untransfused patients with β-thalassemia may require iron chelation to avoid the morbidity and mortality associated with severe iron overload.¹  Using a mouse model of untransfused β-thalassemia (Hbb^th3/+), we previously showed that anemia, ineffective erythropoiesis, and iron overload were ameliorated by increasing hepcidin and causing iron restriction, yielding diminished hemichrome formation and decreased erythroid cell damage.^2,3  Another disease in which iron restriction is known to be beneficial is polycythemia vera (PV). PV is a myeloproliferative disorder triggered by activating mutations in JAK2, an important kinase in the principal signaling pathway of the erythropoietin (EPO) receptor.^11-15  Most of the clinical manifestations of PV are caused by an increased erythrocyte count, leading to an increased risk of pulmonary hypertension and thrombosis. The mainstay of therapy for PV is therapeutic phlebotomy, the goal of which is to reduce the hematocrit (HCT) level to ≤45% to minimize the risk of thrombosis.¹⁶  Because phlebotomy does not influence the underlying defect in bone marrow (BM) red blood cell (RBC) production, the effect of phlebotomy is transient, until patients become iron deficient. As alternative approaches, evidence suggests that systemic iron deficiency or erythroid-targeted iron restriction could be effective in PV patients via inhibition of erythropoietin signaling downstream of JAK2.^11-15  Minihepcidins are small engineered peptides that reproduce the iron-restrictive effects of the hormone hepcidin.^17,18  In a mouse model of severe hemochromatosis (Hamp-KO), administration of a minihepcidin prevented iron overload and caused partial redistribution of iron from the liver to the spleen.^17,18  Minihepcidins caused dose-dependent inhibition of intestinal iron absorption and, at higher doses, restricted the iron supply to erythropoiesis. Based on these observations, this study investigates whether minihepcidin ameliorates anemia and iron overload in β-thalassemic mice and prevents erythrocytosis in PV mice.

All animals (C57BL/6 background) were bred at the mouse facility of Weill Cornell Medical College and Children’s Hospital of Philadelphia. The mice received ad libitum access to water and normal chow that contained 200 ppm iron. We used Hbb^th3/+ mice (The Jackson Laboratory) as a model of β-thalassemia intermedia. An animal model of PV was obtained by crossing mice carrying a Jak2^V617F conditional knockin allele with mice expressing Cre recombinase under the control of VAV regulatory elements (VAV-Cre).¹¹ Jak2^V617F/+ VAV-Cre double-transgenic mice developed a PV-like phenotype, which was transplantable, as previously demonstrated.¹¹  To determine if treatment with minihepcidin could improve hypoxia in our mouse model of β-thalassemia, the hypoxia reporter mouse, oxygen-dependent degradation luciferase (ODD-Luc), was crossed with Hbb^th3/+ and in vivo luminescence imaging was performed at the end point.⁵  All animals were treated with either vehicle control or minihepcidin (M004 or M009) by subcutaneous (SC) injections twice a week for the indicated duration. Hbb^th3/+ animals were also exposed to the oral iron chelator deferiprone (DFP) alone or in combination with minihepcidin. Animals were sacrificed 3.5 days after the final dose for analyses. Blood samples were analyzed as previously described.³ 

The minihepcidins M004 and M009 were prepared fresh for each administration. Each drug was formulated in ethanol 100%, SL-220 (SUNBRIGHT) and water at desired concentration. The ethanol concentration was 10% of the final volume injected. Minihepcidins were administrated twice weekly for 6 weeks, and data were collected 3.5 days after the last injection. The iron chelator deferiprone (Sigma-Aldrich) was added into the drinking water at a concentration of 1.25 mg/mL.

Serum parameters (iron, transferrin saturation [TSAT]) were measured using the Integra 800 Automated Clinical Analyzer (Roche Diagnostics) or the Iron/TIBC Reagent Set (BioPacific Diagnostic). Serum EPO was analyzed using the Mouse Erythropoietin Quantikine ELISA Kit (R&D Systems) following the manufacturer’s instructions. The level of serum hepcidin was determined using the Hepcidin-Murine-Compete kit (Intrinsic LifeSciences) following manufacturer’s instructions. Tissue iron content for Hbb^th3/+ was assayed at the Diagnostic Center for Population and Animal Health at Michigan State University (Lansing, MI).

Erythropoiesis and ROS in BM and spleen cells were analyzed as already described.²  For all analyses, cells were sorted using a FACSCalibur instrument (BD Biosciences) and the results analyzed with FlowJo software (Tree Star).

To visualize membrane-bound globins (hemichrome), we used urea gel electrophoresis (TAU gel), which separates α- and β-globin subunits using denaturing conditions. The same number of erythrocytes was processed for each sample that was loaded onto the gel (150 × 10⁶).² 

RBCs of Hbb^th3/+ mice are characterized by shortened lifespan compared with wild-type (WT) mice. To evaluate whether treatment with minihepcidin could improve this feature, we performed lifespan analysis of peripheral RBCs using flow cytometry as previously described.³ 

Total RNA from livers was purified using the PureLink RNA Mini Kit (Ambion, Life Technologies) and analyzed with the SuperScript III Platinum SYBR Green One-Step qRT-PCR Kit (Invitrogen, Life Technologies). Messenger RNA concentrations of the target genes were normalized by a housekeeping gene (GAPDH).

ODD-Luc-Hbb^th3/+ mice were treated for 6 weeks with biweekly injections of M004 or vehicle. At the end point, in vivo imaging was performed. Animals received luciferin substrate by IV injection. Images were captured using the Xenogen IVIS Optical Imaging System (Xenogen). The system specifically and quantitatively images the expression of the reporter luciferase (Luc) gene in mice.

A sandwich enzyme-linked immunosorbent assay (ELISA) using mouse monoclonal antibodies against mouse erythroferrone (Erfe) was performed as described by Kautz et al.¹⁹ 

Data are presented as mean ± standard deviation (SD). When multiple comparisons were needed, statistical analysis was performed using one-way analysis of variance (ANOVA) with Tukey multiple comparison adjustment. For comparisons between 2 groups, we used a 2-tailed Mann-Whitney U test. All data were analyzed using GraphPad Prism version 6 (Microsoft GraphPad Software, La Jolla, CA).

All animal studies were conducted under protocols approved by the Institutional Animal Care and Use Committee of Weill Cornell Medical College and Children’s Hospital of Philadelphia.

Studies were performed with 2 minihepcidin peptides, designated M004 and M009. M004 is a potent analog of the 9 N-terminal amino acids of hepcidin. M004 reduces cell-surface expression of ferroportin with a 50% effective concentration of 9.7 nM and causes >80% reduction in serum iron 24 hours after SC administration of a dose of 7.5 mg/kg in rats. M004 was discontinued in favor of M009, which was more stable in aqueous solution and therefore easier to administer. M009 is a prodrug that is converted to M004 following systemic administration. Chemically, M009 is the same as M004, except for the addition of an S-methyl group on Cys7, and because the iron-restrictive effect of M009 is determined by the generation of M004, we consider them to be interchangeable treatments. Administration of M009 at a dose of 7.5 mg/kg generates high levels of M004 and a >80% reduction in serum iron that was sustained for at least 48 hours (supplemental Figure 1, available on the Blood Web site).

Two-to 3-month-old Hbb^th3/+ mice are anemic and manifest organ iron overload but do not require RBCs transfusions.^20-22  In order to test whether minihepcidin could improve both iron overload and anemia in young animals, M004 was administered to 2-month-old Hbb^th3/+ male mice. Initial dose-finding studies were conducted using a daily intraperitoneal injection for 2 to 6 weeks of 0.25 to 5 mg/kg per day. In these studies, a dose-dependent decrease in TSAT was observed, consistent with the results observed in WT animals. The dose-dependent effect of minihepcidin-induced iron restriction on ineffective erythropoiesis in the Hbb^th3/+ mouse was biphasic. At the low end of the tested range (0.25 to 1.25 mg/kg per day), we observed normalization of the ratio between mature and immature erythroid cells (CD44⁺-TER119⁺). This was associated with a 15% increase in RBC number and a marked reduction in reticulocyte count (62%) and spleen size (43%) when compared with Hbb^th3/+ animals that received the vehicle (supplemental Figure 2A-B; supplemental Tables 1 and 2). In contrast, at doses of 2.5 to 5 mg/kg per day, severe iron restriction (as low as TSAT = 6%; supplemental Figure 2C and supplemental Table 3) was detected, leading to iron-restricted erythropoiesis and exacerbation of anemia (hemoglobin [Hb] reduction as low as 2.6 g/dL) (supplemental Figure 2D; supplemental Table 4). At these doses, accumulation of immature erythroid precursors was observed by flow cytometry studies, reflecting the expected erythroid maturation block caused by severe iron restriction of erythropoiesis (supplemental Figure 2E, population 1). Prussian blue staining demonstrated the increased sequestration of iron in splenic macrophages and Kupffer cells (supplemental Figure 2F) as expected after the administration of a hepcidin mimetic. The biphasic effect of minihepcidin can be attributed to the fact that at very high doses, the salutary effect of minihepcidin on thalassemic erythropoiesis is outweighed by the inhibitory effect of very severe iron restriction.

Pharmacokinetic data (supplemental Figure 1) indicated that the circulating concentration of active minihepcidin peptides remained high 24 hours after administration and that daily administration would result in accumulation of peptide, potentially leading to severe iron restriction. We therefore explored whether less frequent dosing could achieve sustained improvement in anemia and normalization of tissue iron with diminished risk of iron-restricted erythropoiesis. The dose of 2.625 mg/kg administered subcutaneously twice weekly was selected to generate blood concentrations similar to daily dosing of 0.63 to 1.25 mg/kg per day. The longer dose interval in these studies was chosen to permit the iron-restrictive effect of each dose to dissipate before the next injection and thus avoid severely iron-restricted erythropoiesis.

Two-month-old Hbb^th3/+ males were treated for 6 weeks, twice weekly, with M004 or vehicle control by SC injections. Administration of M004 was associated with an increase in RBCs to levels observed in WT animals (Figure 1A) and an increase in Hb concentration of >2 g/dL (Figure 1B). Reduced spleen weight (−52%; Figure 1C) and reticulocyte count (−60%; Figure 1D) reflected improved erythropoietic efficiency, which was further corroborated by flow cytometry studies. Minihepcidin treatment reduced hemichrome and ROS formation (Figure 1E-F) and improved RBC morphology and lifespan (Figure 1G-H). Hematological improvement was likely caused by enhanced erythroid cell survival resulting from decreased oxidative damage. Total organ iron accumulation was decreased by 50% in the liver and 46% in the kidney (Figure 1I-J). Splenic iron concentration increased by 24%, reflecting retention of iron in splenic macrophages as a result of increased hepcidin activity (Figure 1K). However, total iron levels in the spleen were decreased by 42% (Figure 1L).

Hbb^th3/+ mice were crossed with the hypoxia reporter mouse, ODD-Luc.⁵  In vivo luminescence imaging of these animals showed that mice affected by β-thalassemia are systemically hypoxic as a result of inadequate tissue oxygenation (hypoxia).⁵  ODD-Luc-Hbb^th3/+ mice exhibit increased luminescence (indication of increased hypoxia) mostly in the abdomen and peripheral areas (such as the paws and tails) when compared with nonthalassemic mice.

We postulated that minihepcidin administration could reduce tissue hypoxia by improving tissue oxygenation as a result of increased Hb levels. ODD-Luc-Hbb^th3/+ mice were treated for 6 weeks with 2.625 mg/kg M004 using biweekly SC injections. As shown previously, thalassemic mice treated with minihepcidin exhibit increased erythrocyte number and Hb concentration (Figure 2A-B), decreased reticulocytosis, and reduced splenomegaly (Figure 2C-D). Ineffective erythropoiesis was also ameliorated, as shown by flow cytometry analysis (Figure 2E). At the end point, the animals received IV injection of a luciferin substrate followed by in vivo imaging. Analysis of the results revealed a significant reduction in luminescence mostly in the abdomen of the mice that received minihepcidin compared with control mice (Figure 2F), indicating reduced hypoxia due to improved tissue oxygenation.

In clinical settings, the treatment of β-thalassemia in patients with severe iron overload would likely employ a combination of minihepcidin with oral chelation therapy. We therefore conducted studies in 4- to 5-month-old male Hbb^th3/+ mice with established iron overload to evaluate whether the concurrent use of the oral iron chelator DFP and minihepcidin would cause any therapeutic interactions. M009, which generates M004 in circulation, was administered subcutaneously at the same dose used in previous studies with M004 (2.625 mg/kg). We first analyzed the effect of a single injection of M009. This was sufficient to reduce serum hepcidin up to 36 hours (supplemental Figure 3A), but it had no effect on serum Erfe concentrations (not shown). In this setting, the endogenous hepcidin concentration reached baseline levels at 48 to 72 hours.

M009 or vehicle control was then administered by SC injections biweekly for 6 weeks. Mice were stratified to receive free access to regular water or water containing DFP (1.25 mg/mL). At end-point analysis, iron-overloaded animals treated with DFP alone showed significantly decreased liver iron content (Figure 3A). This was accompanied by a trend toward increased TSAT (Figure 3B) and serum iron (supplemental Figure 3B), suggesting that DFP by itself does not restrict iron delivery to the erythron. The combination of DFP and M009 produced the same reduction in liver iron content as seen with DFP alone. Levels of serum iron (supplemental Figure 3B) and TSAT (Figure 3B) in animals that received M009 alone were reduced compared with the DFP-treated animals and controls, but no differences were detected when the 2 drugs were administrated simultaneously. No differences in serum hepcidin concentrations (supplemental Figure 3C), hepcidin messenger RNA expression (not shown), or expression of other iron-related genes (supplemental Figure 3D-F) were observed between groups. However, ELISA assays of serum Erfe indicated a significant reduction of this factor (Figure 3C).

Treatment with the minihepcidin M009 alone or in combination with DFP produced the same profile of hematological changes described following treatment with M004, resulting in a significant increase in circulating Hb (Figure 4A), RBC count (Figure 4B), and HCT level (Figure 4C). No differences in hematological parameters were observed between DFP-treated and vehicle-treated mice (Figure 4A-C). This indicates that treatment with the iron chelator alone was not sufficient to improve erythropoiesis, despite reduced organ iron content.²³  Reticulocyte count and spleen size were both reduced with M009 and M009 + DFP treatment, reflecting improved erythropoietic efficiency (Figure 4D-E). Flow cytometry studies of BM and spleen erythroid populations from minihepcidin-treated animals demonstrated an increase in the relative proportion of mature erythroid cells (Figure 2E) and reduced levels of ROS (Figure 4F). Parameters of RBC damage (red cell distribution width) (Figure 4G), hemichrome formation (Figure 4H), and erythrocyte morphology (Figure 4I) were improved. We previously demonstrated that Hbb^th3/+ animals show increased EPO concentrations in response to anemia.^20,24  Overexpression of EPO leads to increased phosphorylation of the Jak2 kinase and overexpression of the genes downstream of the Jak2-Stat5 pathway involved in cell proliferation and survival.^20,24,25  As minihepcidin improved erythropoiesis and ameliorated anemia, EPO concentrations declined (Figure 4J).

M009 was also used to treat 4-month-old females Hbb^th3/+ and showed a trend similar to that observed in males. We observed improved liver iron content, hematological parameters, and erythroid maturation, mostly in the spleen, as well as a reduction in splenomegaly (supplemental Figure 4A-F).

The progeny of floxed Jak2^V617F/+ mice crossed with a mouse hemizygous for the Vav-cre transgene expressed in hematopoietic cells can develop PV, and this phenotype can be easily propagated by engrafting Jak2^V617F/+-Vav-cre BM cells into WT C57BL/6 mice.²⁶  Using these mice, we tested the hypothesis that minihepcidin-induced iron restriction could work as medical phlebotomy by reversing erythrocytosis and normalizing the HCT level in murine PV. To this end, we studied the effects of markedly higher doses of M009 than were previously used in studies in the Hbb^th3/+ mice to ensure that a high level of iron restriction (serum iron reduction of >80% throughout the interval between doses) was present throughout the study.

We treated PV mice with vehicle or M009 (10 and 15 mg/kg) administered by biweekly SC injections for 3 weeks, with end-point analysis performed 3.5 days after the last injection. Treatment was initiated 1 month after BM transplantation, when the PV phenotype had been fully established. At 3 weeks, treatment showed a dose-dependent reduction (or normalization) in the HCT level, RBC count, and Hb concentration in animals receiving minihepcidin when compared with mice exposed to the vehicle (Figure 5A-C). Splenomegaly was reduced (Figure 5D), and flow cytometric profiles of erythroid precursors improved (Figure 5E). Flow cytometry analysis indicated that iron restriction mediated by minihepcidin led to a reduction of erythroid progenitor cells when compared with vehicle (fraction 1 to 4). No significant changes in mean corpuscular hemoglobin were seen. We did not observe differences in liver iron content in treated animals vs controls, but a significant increase in splenic iron levels was detected at both doses (supplemental Figure 5A). These findings were confirmed by Perls’ Prussian blue staining in liver and spleen sections (supplemental Figure 5B), showing increased iron retention by macrophages.

To investigate the effect of minihepcidin at similar or higher doses and for longer periods, we performed studies using M009 at a dose range of 10 to 20 mg/kg administered by twice-weekly SC injections for 6 weeks. At the end of the treatment period, we observed decreased HCT levels (Figure 6A) due to a large reduction in erythrocytosis in M009-treated animals. Circulating Hb concentrations and RBC numbers (Figure 6B-C) were also reduced in a dose-dependent fashion to the same or lower level than seen in healthy untreated WT littermates. However, we also observed a dose-dependent development of iron-restricted erythropoiesis by splenomegaly (Figure 6D) and expansion of early erythroid progenitors by flow cytometry studies (Figure 6E). These observations indicate that in PV minihepcidin administration needs to be titrated to reduce erythrocytosis while minimizing the negative effects of erythrocyte iron deficiency. As seen in animals treated for 3 weeks, administration of minihepcidin led to increased iron deposition in splenic macrophages, while no major changes were observed in liver sections (Figure 6F).

In β-thalassemia and anemia, increased EPO levels and expanded but ineffective erythropoiesis result in lowered hepcidin levels, increased intestinal iron absorption, and TSAT.^3,6,8,21,27  In turn, excess iron contributes to the development and worsening of ineffective erythropoiesis and reduced erythroblast maturation.¹  Here, we show that minihepcidins can break this maladaptive, vicious circle and thus establish a new set point for BM erythropoiesis, ameliorating ineffective erythropoiesis and decreasing erythroid cell damage. Remarkably, the RBC count and lifespan are restored to those observed in WT mice. We surmise that the corrective effect begins with a reduction in hemichrome and ROS formation, leading to subsequent amelioration of erythroid damage and increased effective erythropoiesis. We expect that in a clinical setting, these biological effects may result in other beneficial changes not explored in our mouse model. For instance, improved RBC morphology may reduce stickiness, peripheral RBC destruction, and thrombotic risk. Decreased spleen size could reduce the need for surgical splenectomy. An improvement in anemia reduces hypoxia, which, together with decreased destruction of RBCs, may prevent or delay lethal complications such as pulmonary hypertension. Preventing iron accumulation addresses a critical mechanism of the disease and should avoid or delay the development of cirrhosis and liver cancer.

Here, we also conducted studies in the Hbb^th3/+ mouse to evaluate whether the concurrent use of the oral iron chelator DFP and minihepcidin would lead to therapeutic interaction. Our data show that minihepcidin and DFP retain their positive effects on erythropoiesis and iron overload. Based on these observations, we anticipate that in untransfused β-thalassemic patients with moderate or severe iron overload, treatment with minihepcidin could improve erythropoiesis, whereas oral chelation therapy would still be beneficial in reversing established iron overload.

Interestingly, chelation alone failed to cause iron restriction or improve hematological parameters, as also seen in previous studies.^2,28  In DFP-treated mice, despite a decrease in liver iron concentration, TSAT and serum iron levels were not decreased. However, when DFP was combined with minihepcidin, we observed the same beneficial effects on ineffective erythropoiesis, RBC lifespan, and anemia, even though levels of serum iron and TSAT were not decreased compared with solvent-treated mice. The potential mechanisms by which this may occur are speculative. Perhaps as DFP releases iron from the liver, it delivers some iron to serum transferrin via a shuttling effect. However, when minihepcidin is administered together with DFP, the overall flow of iron for erythropoiesis might be reduced, as some iron is diverted to macrophages (supplemental Figure 6). Additional studies will be needed to test this model. Nevertheless, the lack of an impact of concurrent minihepcidin therapy on the tissue iron–depleting properties of DFP indicates that the 2 modalities may have additive or synergistic therapeutic effects. In addition, these data indicate that chelator-induced iron excretion can be accomplished despite increased macrophage iron sequestration in minihepcidin-treated mice. These results also bode well for the design of clinical trials in which the therapeutic effects of minihepcidin should be testable without interference with chelator therapy, the current standard of care. In short-term experiments, endogenous hepcidin levels were reduced, reaching baseline levels at 48 to 72 hours (supplemental Figure 3A). This, together with the fact that endogenous hepcidin levels were not increased at the end of the 6-week study (supplemental Figure 3C) while serum Erfe levels were decreased, points to the multiple regulators of hepcidin, where the net effect of decreasing both Erfe and systemic iron levels resulted in unchanged hepcidin levels. Altogether these observations document the amelioration of ineffective erythropoiesis as indicated by a better balance between immature and mature erythroid cells in the BM and spleen and more RBCs in circulation but less serum EPO (Figures 2E, 4B, and 4J).

Our data also indicate that minihepcidin may be beneficial in the treatment of PV by limiting iron availability to erythroid precursors, thereby inhibiting JAK2-stimulated erythropoiesis and reducing RBC production. Controlling HCT is of critical importance in PV management. Here, the ability of minihepcidin to rapidly impose iron-restricted erythropoiesis contrasts with intermittent phlebotomy, where the removal of RBCs is rapidly compensated, at least until enough RBC mass is removed to cause iron deficiency. These results suggest that minihepcidin therapy could be used as a “medical phlebotomy” to provide continuous control of accelerated erythropoiesis in the treatment of PV. Because of the association between high HCT levels and poor outcomes, the rapid and continuous control of HCT levels using minihepcidin could be highly clinically meaningful if clinical trials show that it can be done safely.

In conclusion, the drug-like minihepcidin peptides offer a new potential pharmacotherapeutic approach to achieve therapeutic iron restriction in the treatment of β-thalassemia, PV, and perhaps other less common disorders associated with abnormal erythropoiesis.

This work was supported by Merganser Biotech and grants from the National Institute of Diabetes and Digestive and Kidney Diseases and National Heart, Lung, and Blood Institute of the National Institutes of Health: R01 DK095112 and R01 DK090554 (S.R.), DK095201 (Y.M.S.), R01 DK090554 (T.G. and E.N.), R01 DK107309 (E.N.), R01 DK107670 (Y.G.), R01 DK095112 (R.E.F., S.R., and Y.G.), and K08 HL105682 (Y.G.). C.C. is supported by the Cooley's Anemia Foundation, and P.R.O. is supported by The Child Reach Foundation.

Contribution: C.C., P.R.O., H.C., Y.G., P.P., and E.V.V. performed research and analyzed the results; V.N. provided support with statistical analysis; Y.G., R.E.F., Y.M.S., E.N., and T.G. discussed the experiments and reviewed the paper; and S.R., C.C., and B.M. designed research, analyzed results, and wrote the paper.

Conflict-of-interest disclosure: B.M. has stock in Merganser Biotech. S.R. has restricted stocks in Merganser Biotech. S.R. is a consultant for Novartis Pharmaceuticals, Bayer Healthcare, Merganser Biotech and Keryx Pharmaceuticals. S.R. is a member of scientific advisory board of Merganser Biotech and Ionis Pharmaceuticals. T.G. and E.N. are consultants and stockholders of Intrinsic LifeSciences, Merganser Biotech, and Silarus Therapeutics. The remaining authors declare no competing financial interests.

Correspondence: Stefano Rivella, Children’s Hospital of Philadelphia, Abramson Research Center, 3615 Civic Center Blvd, Room 316B, Philadelphia, PA 19104; e-mail: rivellas@email.chop.edu.