Skip to main content
PMC home page
eLife logoLink to eLife
. 2022 Jan 11;11:e73456. doi: 10.7554/eLife.73456

Aging is associated with increased brain iron through cortex-derived hepcidin expression

Tatsuya Sato 1, Jason Solomon Shapiro 1, Hsiang-Chun Chang 1, Richard A Miller 2, Hossein Ardehali 1,✉
Editors: Yelena Ginzburg3, Mone Zaidi4
PMCID: PMC8752087  PMID: 35014607

Abstract

Iron is an essential molecule for biological processes, but its accumulation can lead to oxidative stress and cellular death. Due to its oxidative effects, iron accumulation is implicated in the process of aging and neurodegenerative diseases. However, the mechanism for this increase in iron with aging, and whether this increase is localized to specific cellular compartment(s), are not known. Here, we measured the levels of iron in different tissues of aged mice, and demonstrated that while cytosolic non-heme iron is increased in the liver and muscle tissue, only the aged brain cortex exhibits an increase in both the cytosolic and mitochondrial non-heme iron. This increase in brain iron is associated with elevated levels of local hepcidin mRNA and protein in the brain. We also demonstrate that the increase in hepcidin is associated with increased ubiquitination and reduced levels of the only iron exporter, ferroportin-1 (FPN1). Overall, our studies provide a potential mechanism for iron accumulation in the brain through increased local expression of hepcidin, and subsequent iron accumulation due to decreased iron export. Additionally, our data support that aging is associated with mitochondrial and cytosolic iron accumulation only in the brain and not in other tissues.

Research organism: Mouse

Introduction

Iron is an essential molecule for almost every organism on earth. It is generally found naturally in two oxidation states, either as the oxidized form of ferric iron (Fe3+), or as the reduced form of ferrous iron (Fe2+). Ferrous iron is the major form of iron used in mammalian cells, and due to the scarcity of the reduced form of iron on earth, mammalian cells have developed mechanisms for efficient uptake of oxidized iron through its solubilization by acidification, followed by reduction and its cellular transportation. Iron can donate and accept electrons from various substrates due to its unique oxidation-reduction properties, making it an important cofactor in mammalian cells. Iron is also essential for heme and Fe-S clusters and exerts other biological effects through its role in processes such as demethylation, dehydrogenation, and reduction of sulfur (Koleini et al., 2021).

Due to its oxidative effect, iron accumulation is hypothesized to lead to oxidative stress and cellular damage that accelerates the process of aging (Xu et al., 2012). Aging has been shown to be associated with iron accumulation in various organs and tissues both in humans and in animal models of aging (Xu et al., 2012). However, whether iron accumulation occurs in all organs or is specific to one organ is not known. In the brain, iron accumulation has been demonstrated both in animal models of aging (Benkovic and Connor, 1993; Hahn et al., 2009; Roskams and Connor, 1994) and in humans (Bartzokis et al., 2010; Bartzokis et al., 1994a; Bartzokis et al., 1994b; Zecca et al., 2001). Additionally, it is now believed that iron accumulation may be the cause of neuronal cell death in some of neurodegenerative diseases (Lei et al., 2012). Despite the importance of iron in brain disorders, the mechanism by which iron accumulates in the brain with aging is not known. Previous studies suggest that iron homeostasis in the brain is dependent on normal expression of genes involved in iron uptake into the cells and its cellular storage and regulation (Qian and Wang, 1998). Another study demonstrated that only one protein involved in the iron import pathway, divalent metal transporter (DMT1) was altered in aged brain, while the levels of key proteins involved in iron import and export, transferrin receptor-1 (TfR1), and ferroportin-1 (FPN1) were unchanged (Lu et al., 2017). However, the mechanism for the age-related increase in iron accumulation in the brain remains unclear.

In this paper, we assessed the levels of iron in different tissues and cellular compartments with aging in mice. Since it has been reported that age-related brain atrophy, which is associated with age-related cognitive decline, is evident in the brain cortex (Nyberg et al., 2010), we selected the cerebral cortex as the brain region in the present study. We demonstrate that the cytosolic non-heme iron is increased in the muscle and liver tissue, while the brain cortex has increased non-heme iron both in the mitochondria and in the cytosol. We also demonstrate that hepcidin levels are upregulated in the brain cortex of aged mice, which is associated with ubiquitination of FPN1 and a reduction in its protein levels. Thus, our studies suggest that local expression of hepcidin in the brain may be a driver in increasing iron levels due to a reduction in iron export by FPN1 degradation.

Results

Iron is increased in the brain cortex with aging

To determine whether iron accumulates in tissues with aging, we measured heme and non-heme iron levels in the mitochondria and cytoplasm of liver, skeletal muscle, and brain tissues. These organs were selected since liver is the main iron storage organ, while skeletal muscle is a major iron-consuming organ. Additionally, brain iron accumulation is reported with neurodegenerative diseases. We chose two time points for our studies: 4 months for young mice and 22 months for old mice. The almost 2 years age of the old mice is considered sufficient to result in senescence of some organs (Flurkey et al., 2007). Our data demonstrated that cytosolic non-heme iron in the liver was increased with age, whereas liver iron contents in the mitochondria, where iron accumulation is associated with cell dysfunction and cell death, were unchanged with age for both heme and non-heme forms (Figure 1A and B). Similarly, we observed an increase in the cytosolic, but not mitochondrial, non-heme iron in the gastrocnemius muscle. We also did not see a change in the cytosolic heme iron; however, we were not able to detect heme iron in the mitochondria of skeletal muscle tissue, possibly due to technical limitations associated with the isolation of intact mitochondria from skeletal muscle tissue (Figure 1C and D).

Figure 1. Non-heme iron is increased in the cytosol and mitochondria of the brain cortex with aging.

Figure 1.

Open in a new tab

Liver non-heme (A) and heme (B) iron in young (4 months old) and old (22 months old) mice from the cytosol and mitochondria (n=6). Gastrocnemius non-heme (C) and heme (D) iron in young and old mice from the cytosol and mitochondria (n=6). Outliers in non-heme iron in the cytosol (n=1) and in the mitochondria (n=2) were excluded. Mitochondrial heme iron in gastrocnemius muscle was not detectable in our studies. Brain non-heme (E) and heme (F) iron in young and old mice from the cytosol and mitochondria (n=6). (G) Immunoblots of mitochondrial and cytosolic markers, confirming the purity of mitochondrial isolation in the brain cortex tissue. W = whole cell lysate, Cyt = cytosolic fraction, Mit = mitochondrial fraction, SDHA = Succinate Dehydrogenase Complex Flavoprotein Subunit A, HPRT = hypoxanthine phosphoribosyltransferase. (H) Immunoblots of FTN and IRP2 in the brain of young and aged mice (n=4). FTN = ferritin, IRP2 = iron regulatory protein 2, H = heavy chain, L = light chain. Densitometric quantification of light and heavy chain FTN (I) and IRP2 (J) in the brain of young and aged mice. * p<0.05.

Figure 1—source data 1. Full-length images of immunoblots shown in Figure 1.
Figure 1—source data 2. Iron measurement – Original data of iron measurements shown in Figure 1.
Figure 1—source data 3. Densitometry – Original data of densitometric quantifications of immunoblots shown in Figure 1.

We then studied iron levels in the brain mitochondrial and cytosolic fractions, and noted a significant increase in non-heme iron in both the mitochondria and cytoplasm of dissected brain cortex in older mice (Figure 1E). However, there was no increase in heme iron levels in the either compartment in the brain cortex (Figure 1F). The purity of the mitochondrial fraction in the brain cortex tissue was confirmed by immunoblotting (Figure 1G). Consistent with increased non-heme iron levels, we also observed increased protein levels of ferritin light and heavy chains (Figure 1H and I). This increase may be a cellular protective mechanism because free or loosely bound iron in the form of non-heme iron is detrimental to the cells, and iron stored in ferritin is considered redox-inert. Additionally, an elevated level of cellular iron leads to the degradation of IRP2 protein (Wang et al., 2004). We therefore measured IRP2 protein levels in the brain cortex and observed a decrease in IRP2 protein levels in the brain cortex of aged mice (Figure 1H and J), consistent with an increase in iron levels with aging.

Hepcidin levels are increased in the brain with aging

Since non-heme iron is increased in both the cytosolic and mitochondrial compartments of the brain cortex and previous reports suggest a potential role for iron in neurodegenerative diseases associated with aging such as Alzheimer’s and Parkinson’s disease (Berg et al., 2001; Smith et al., 1997), we then focused our studies on the mechanism for the increase in iron in the brain tissue. We measured the mRNA levels of all proteins involved in cellular iron homeostasis. We conducted these studies both in the tissues of liver and brain cortex. In the liver, we noted a significant increase in the mRNA levels of Fpn1, heme oxygenase 1 (Hmox1), ALA dehydrogenase (Alad), and bone morphogenetic protein 6 (Bmp6), and a significant decrease in the mRNA level of Metalloreductase Six-Transmembrane Epithelial Antigen of Prostate-3 (Steap3) (Figure 2A). In contrast, we noted a significant increase in the mRNA levels of TfR2, mitoferrin 2 (Mfrn2), tristetraprolin (Ttp), Hmox1, Alad, and hepcidin (Hamp1) in the brain cortex of aged mice (Figure 2B). TfR2 does not play a meaningful role in iron uptake compared to transferrin-receptor protein (TfRC), since it has a much lower affinity for transferrin (Kawabata et al., 1999). Instead, it acts as a signaling molecule to regulate Hamp1 transcription through SMAD signaling (Silvestri et al., 2014). Since both Tfr2 and Hamp1 are increased in the brain cortex with aging, we then focused our studies on the cellular hepcidin pathway. To confirm that the increase in Hamp1 mRNA with aging is associated with an increase in its protein, we measured HAMP1 protein in the tissues of brain cortex in young and aged mice and noted a significant increase in its protein levels (Figure 2C and D). Finally, we performed immunohistochemistry on the brain frontal cortex of aged mice and confirmed that HAMP1 protein levels are significantly increased with aging (Figure 2E).

Figure 2. Hepcidin protein expression is significantly increased in the aged brain cortex.

Figure 2.

Open in a new tab

mRNA levels of proteins involved in iron regulation in the liver (A) and brain (B) in young (4 months old) and old (22 months old) mice (n=10/group). Means and SEM are indicated as horizontal and vertical bars, respectively. An undetected measurement (n=1) in the brain Fech and undetected measurements (n=2) and an outlier (n=1) in the brain Hamp1 were excluded. Tfr1 = transferrin receptor 1, Tfr2 = transferrin receptor 2, Dmt1 = divalent metal transporter 1, Fpn1 = ferroportin 1, Pcbp1 = Poly(RC) Binding Protein 1, Steap3 = Metalloreductase Six-Transmembrane Epithelial Antigen Of Prostate 3, Sec15l1 = exocyst complex component 6, Mfrn2 = mitoferrin 2, Abcb7 = ATP-binding cassette sub-family B member 7, Abcb8 = ATP-binding cassette sub-family B member 8, Alr1 = augmenter of liver regeneration, Cp = ceruloplasmin, Fxn = frataxin, Ttp = tristetraprolin, Lias = tristetraprolin, Hmox1 = heme oxygenase 1, Hmox2 = heme oxygenase 2, Alas1 = 5′-aminolevulinate synthase 1, Alas2 = 5′-aminolevulinate synthase 2, Fech = ferrochelatase, Alad = aminolevulinate dehydratase, Pbgd = porphobilinogen deaminase, Uros = uroporphyrinogen III synthase, Urod = uroporphyrinogen decarboxylase, Ppox = protoporphyrinogen oxidase, Abcb10 = ATP-binding cassette, sub-family B member 10, Hamp1 = hepcidin1, Bmp6 = bone morphogenetic protein 6, Hfe = homeostatic iron regulator, Ftmt = mitochondrial ferritin. (C) Representative immunoblot for hepcidin1 in the brain (n=6). (D) Summary of densitometry analysis of panel (C). (E) Representative immunohistochemistry of hepcidin1 (Red = anti-hepcidin1, blue = DAPI) in the brain frontal cortex of young and aged mice. Scale bar=200 µm. * p<0.05.

Figure 2—source data 1. Full-length images of immunoblots shown in Figure 2.
Figure 2—source data 2. qRT-PCR liver – Original data of qRT-PCR in the liver shown in Figure 2.
Figure 2—source data 3. qRT-PCR brain – Original data of qRT-PCR in the brain shown in Figure 2.
Figure 2—source data 4. Densitometry – Original data of densitometry quantifications of immunoblots shown in Figure 2.

Aging is associated with decreased FPN1 protein levels through its ubiquitination

The major effect of hepcidin protein on iron regulation is through ubiquitination and subsequent degradation of FPN1, resulting in a decrease in cellular iron export (Qiao et al., 2012). This mechanism of hepcidin effect generally occurs in intestinal cells and macrophages, which leads to decreased iron export from these cells. Since HAMP1 protein levels are increased in the brain cortex of aged mice, we hypothesized that FPN1 levels would be decreased in the brains of these mice. FPN1 protein levels are significantly decreased in the brain cortex tissue of aged mice, while TfR1 levels are unchanged (Figure 3A and B). Since mRNA levels of FPN1 are not significantly changed in the brain cortex, we hypothesized that the decrease in FPN1 protein might be due to its degradation by hepcidin. We therefore measured FPN1 protein ubiquitination in the brains of aged mice, and showed a significant increase in its ubiquitination (Figure 3C and D). These results indicate that FPN1 protein is ubiquitinated and decreased with aging, perhaps as a consequence of the higher levels of hepcidin.

Figure 3. FPN1 protein level is decreased while its poly-ubiquitination is increased in the brain cortex of aged mice.

Figure 3.

Open in a new tab

(A) Immunoblots of iron transporting proteins TfR1 and FPN1 in the brain cortex of young and aged mice (n=3). (B) Summary of densitometric analysis of panel (A). (C) Poly-ubiquitination levels of FPN1, as assessed by immunoprecipitation, in the brain cortex of young and aged mice (n=3). (D) Summary of the densitometric analysis of panel (C). (E) Fe-S cluster containing aconitase enzyme activity, a marker of cellular oxidative stress, in the brain cortex of young and aged mice (n=4). c-aconitase = cytosolic aconitase (ACO1), m-aconitase = mitochondrial aconitase (ACO2). * p<0.05.

Figure 3—source data 1. Full length images of immunoblots shown in Figure 3.
Figure 3—source data 2. Densitometry – Original data of densitometry quantifications of immunoblots shown in Figure 3.
Figure 3—source data 3. Aconitase assay – Original data of aconitase enzyme activities shown in Figure 3.

We then assessed whether the increase in iron in the brain cortex of aged mice is associated with increased oxidative stress. Aconitase is a Fe/S-containing protein that is highly sensitive to oxidative stress (Noster et al., 2019). Our results demonstrated a decrease in both cytosolic and mitochondrial aconitase activity in the brain cortex of aged mice (Figure 3E), being consistent with the notion that oxidative stress is increased in the aged brain (Kandlur et al., 2020).

Discussion

Iron is critical for normal cell function, but its excess can lead to cellular damage due to oxidative stress. Thus, the levels of cellular iron, particularly in the mitochondria, are tightly regulated (Koleini et al., 2021). Iron has been implicated in the development of some neurodegenerative diseases that are associated with aging, including Alzheimer’s and Parkinson’s diseases (Xu et al., 2012). However, the molecular mechanism for the increase in brain iron in aging is not known. In this paper, we showed that iron accumulates in the mouse brain cortex with aging and that this is associated with a significant increase in hepcidin production in the brain cortex. Since the major function of hepcidin is to bind to FPN1, leading to FPN1 degradation, we assessed levels of FPN1 in extracts of brain cortex from young and aged mice and showed that while its mRNA is not changed, protein levels of FPN1 are significantly decreased. We also demonstrate that brain FPN1 ubiquitination is increased with aging. Finally, we demonstrate that there is increased oxidative stress in the brains of aged mice, which may well reflect, at least in part, accumulation of iron, perhaps in the context of other age-dependent changes in oxidative damage and defenses against oxidative injury.

Hepcidin antimicrobial peptide (HAMP1) is a small 25–amino acid hormone peptide that is secreted by hepatocytes in conditions of iron sufficiency, and which binds to FPN1 and promotes its internalization and degradation. FPN1 degradation inhibits iron absorption from the digestive tract and the release of iron from macrophages. Although liver is believed to be a major source of systemic hepcidin production, this peptide has recently been shown to be expressed by other tissues. A recent study demonstrated that hepcidin is expressed in the heart and that its cardiac-specific deletion has detrimental effects on cardiac function (Lakhal-Littleton et al., 2016). Additionally, hepcidin has been inferred to play a role in colorectal cancer by sequestering iron to maintain the nucleotide pool and sustain proliferation in colorectal cancer (Schwartz et al., 2021). Recent studies have also demonstrated that hepcidin overexpression in astrocytes protects against amyloid-β induced brain damage in mice and also protects against mouse models of Parkinson’s disease (Liang et al., 2020; Zhang et al., 2020). On the other hand, it has been reported that upregulation of local hepcidin expression is associated with brain iron accumulation in patients with Alzheimer’s disease (Chaudhary et al., 2021). However, it was not known whether hepcidin is increased with changes in its target FPN1 in the aged brain in mice. In the present study, our data show that brain hepcidin is increased with aging and that it likely leads to a reduction in FPN1 levels and iron accumulation. Since the results of this study are based on a small number of immunoblot samples using single-gender of mice, additional studies are needed to confirm these findings in a larger cohort of mice of both genders.

Previous reports have suggested that inflammation plays a role in the brain-derived hepcidin expression via lipopolysaccharide- or interleukin-6-mediated STAT3 pathway (Lieblein-Boff et al., 2013; Wang et al., 2008), and another study suggested that toll-like receptor 4/myD88-mediated signaling of hepcidin expression cause brain iron accumulation and oxidative injury (Xiong et al., 2016). Thus, age-related inflammation may be a regulator of iron accumulation in the brain via increased brain-derived hepcidin expression and subsequent FPN1 downregulation. In contrast, BMP6 secreted by the liver sinusoidal endothelial cells in response to iron binds to BMPR and induces hepcidin production via the SMAD pathway. However, it is not known whether BMP6 regulates hepcidin expression in the brain. Indeed, our data showed that gene expression of BMP6 was not increased in the aged brain. Additionally, TfR2 is associated with positive regulation of hepcidin expression (Silvestri et al., 2014), and mitochondrial iron transport in the brain (Mastroberardino et al., 2009). However, the expression of brain hepcidin is altered in TfR2-deficient mice in response to systemic iron deficiency or iron overload (Pellegrino et al., 2016), suggesting that TfR2 is not the sole regulator of brain hepcidin expression. Further studies are needed to clarify whether increased TfR2 in the aged brain contributes to the increase in brain-derived hepcidin expression or alternatively contributes to iron accumulation with age.

Lu et al., 2017 have shown that, compared to younger animals, 24-month-old rats have significant increase in hepcidin expression in the brain cortex, in addition to hippocampus, striatum, and substantia nigra. However, FPN1 levels were not significantly changed in different brain regions. The authors concluded that increased DMT1, rather than TfR1 or FPN1, may be responsible for the increased brain iron with aging. Although the increase in hepcidin is consistent with our data, their proposed mechanism differs from what we observed, which could be due to differences in species, and sample size between our studies. Additionally, our data showed that mRNA and protein levels of TfR1 do not change with aging. TfR1 is regulated by both iron-dependent and iron-independent mechanisms. For iron-dependent pathways, IRPs are well-known positive regulators of TfR1. However, TfR1 is also regulated by iron-independent immune response, such as NF-kB- and/or HIF-mediated pathways (Samavati et al., 2008). Since we found that decreased IRP2 in the aged brain, it is possible that TfR1 gets downregulated via decreased IRP2-mediated pathway, but iron-independent regulation of TfR1 cancels IRP2-mediated decrease in the aged brain. Our data therefore cannot resolve whether TfR1 is central to age-mediated iron accumulation in the brain and if so, by what mechanism. Additional studies are necessary to clarify this important point.

A recent study reported that in patients with Parkinson’s disease, the activity of heme oxidase-1, which is regulated by HMOX1 and degrades heme, is increased in the serum and brain and its activation is associated with anemia and iron accumulation in the substantia nigra, but not other brain regions (Xu et al., 2021). Considering that cytosolic and mitochondrial heme iron levels were not changed with aging in both liver and brain in the present study, the increase in ALAD expression (which can catalyze heme synthesis) might be a compensatory mechanism against increased degradation of heme via increased HMOX1. However, the changes noted in HMOX1 and ALAD would not explain the increase in non-heme iron in the brain.

The mechanism for iron-mediated aging is likely due to the oxidative effects of iron and cellular damage. Mitochondria are the major site of reactive oxygen species production in the cells, thus, the levels of mitochondrial iron are tightly controlled. In our studies, we noted an increase only in the cytosolic non-heme iron in the liver and the skeletal muscle of aged mice, while non-heme iron was increased in both the mitochondrial and cytosolic fraction of the brain cortex tissue. The significance of sole increase in cytosolic iron is not clear, but the increase in mitochondrial iron in the brain is significant since it likely contributes to the oxidative damage that occurs in this tissue with aging. It is also possible that cytosolic iron accumulation in the liver and muscle tissue is due to mild increased systemic iron. Since the brain is not an iron storage organ and iron movement to the brain has to go through the blood-brain barrier (BBB), iron accumulation in the brain with aging is likely independent of systemic iron metabolism.

Our studies have clinical implications for neurodegenerative diseases associated with aging through two possible mechanisms: (1) by targeting brain iron with iron chelators that can cross the BBB, and (2) through a reduction in locally produced hepcidin. There is currently interest in taking the first approach and clinical studies support this approach (Sun et al., 2018). However, targeting hepcidin in the brain is technically challenging and may require the design of small molecule hepcidin inhibitors that can also cross the BBB. We also acknowledge that we do not identify the specific cells in the brain that express hepcidin and FPN1. Further studies are needed to address cellular specificity and interaction among increased brain-derived hepcidin regulation, Fpn1 expression, and cytosolic and mitochondrial iron accumulation with aging. Identification of these pathways may have therapeutic potentials.

Materials and methods

Key resources table.

Reagent type (species) or resource Designation Source or reference Identifiers Additional information
Strain, strain background(Mus musculus, female, C57BL/6) UM-HET3 Dr. Miller lab Harrison et al., 2021 (cited in the paper)
Strain, strain background(M. musculus, female, C57BL/6) Wild type Jackson Laboratories  
Antibody Rabbit polyclonal anti-HPRT antibody ProteinTech 150-59-1-AP WB (1:5000)
Antibody Rabbit polyclonal anti-SDHA antibody Invitrogen 45-920-0 WB (1:1000)
Antibody Rabbit polyclonal anti-FTN antibody Sigma-Ardrich F5012 WB (1:1000)
Antibody Rabbit polyclonal anti-IRP2 antibody Sigma-Ardrich SAB2101174 WB (1:500)
Antibody Rabbit polyclonal anti-TFR1 antibody ProteinTech 100-84-2-AP WB (1:1000)
Antibody Rabbit polyclonal anti-FPN1 antibody Novus Biologicals NBP1-21502 WB (1:1000)
Antibody Rabbit monoclonal anti-Hepcidin antibody Abcam 190775 WB (1:200)IH (1:20)
Antibody Mouse polyclonal anti-ubiquitin antibody Cell Signaling Technologies 3933S WB (1:1000)
Antibody Mouse monoclonal anti-GAPDH antibody ProteinTech 60004-1-Ig WB (1:5000)
Antibody Rabbit polyclonal anti-α-Tubulin antibody ProteinTech 66031-1-Ig WB (1:5000)
Antibody HRP-conjugated donkey polyclonal anti-rabbit IgG antibody Jackson ImmunoResearch 711-035-152 WB (1:5000)
Antibody HRP-conjugated donkey polyclonal anti-mouse IgG antibody Jackson ImmunoResearch 715-035-150 WB (1:5000)
Antibody Alexa Fluor 594 goat polyclonal anti-rabbit IgG Jackson ImmunoResearch 111-585-144 IH (1:200)
Chemical compound, drug 3-(2-Pyridyl)–5,6-di(2-furyl)–1,2,4-triazine-5′,5″-disulfonic acid disodium salt(Ferrozine) Sigma-Aldrich 82940  
Chemical compound, drug Trichloroacetic acid Sigma-Aldrich T6399  
Chemical compound, drug Thioglycolic acid Sigma-Aldrich T3758  
Chemical compound, drug Hemin Sigma-Aldrich H9039  
Chemical compound, drug RNA-STAT60 Teltest Cs-502  
Chemical compound, drug Glycogen Life Technologies AM9510  
Chemical compound, drug Paraformaldehyde Thermo Fisher Scientific AC416780250  
Chemical compound, drug RIPA Buffer Thermo Fisher Scientific 89901  
Commercial assay or kit ProteaseArrest Protease Inhibitor G-Biosciences 786-437  
Commercial assay or kit qScript cDNA Synthesis Kit Quanta 95047-500  
Commercial assay or kit PerfeCTa SYBR Green FastMix Quanta 95074-05K  
Commercial assay or kit SuperSignal West Pico PLUS Chemiluminescent Substrate Pierce 34579  
Commercial assay or kit Dynabeads Protein G for Immunoprecipitation Invitrogen 10003D  
Commercial assay or kit BCA Protein Assay Kit Pierce 23225  
Commercial assay or kit Mitochondria Isolation Kit for Tissue Pierce 89801  
Commercial assay or kit Aconitase Activity Assay Kit Abcam Ab109712  
Commercial assay or kit OTC compound Sakura Finetek 4583  
Sequence-based reagent Primers for qRT-PCR This manuscript N/A Included in the next table
Software, algorithm GraphPad Prism GraphPad Version 9  
Software, algorithm ImageJ NIH 1.53c  
Open in a new tab

Primer sequences

Genes Forward primer Reverse primer
Actb CCGTGAAAAGATGACCCAGAT GTACATGGCTGGGGTGTTG
Hprt1 CTGGAAAGAATGTCTTGATTGTTG TGCATTGTTTTACCAGTGTCAA
Tfr1 GCATTGCGGACTGTAGAGG GCTTGATCCATCATTCTCAGC
Tfr2 AGCCCATCAGTGCTGACATT AGGAGAGCCTGAGAGGTGAC
Dmt1 GGCGTGTGTGGAGGTGGCGG TGGTCCCCAGAAGCGCCATCG
Fpn1 TGGCCACTCTCTCTCCACTT ACACTGCAAAGTGCCACATC
Pcbp1 AGATCAAAATTGCCAACCCG AGCCAGTAATAGTGACCTGC
Steap3 AGGAGTTCAGCTTCGTGCAG GAGGGCTAGACAAGATGCGTA
Sec15l1 CCACCCTCCGATCTGTGTAT AAACTTCTTGTGTGCGTTTGG
Mfrn2 AGTGACGTAATCCACCCAGGGGGC CTCTGCTTGACGACTTCCGCTGG
Abcb7 GGAGGAGGACTCCACACAGA ATGGCAACTCTGGCTCGTAG
Abcb8 GGGCAACAGGTGTAGCAGAT TGCTTGATAGCGTTCCTCCT
Alr1 CCCTGCGAGGAATGTGCGGAA TCACCTCATTGTGCAGGCGGC
Cp TGCTCCTTCTGGGACGGACATCTTC GCCACCAATTCTTGTGGCACCTTGC
Fxn CAGACAAGCCCTATACCCTG AGCCAGATTTGCTTGTTTGG
Ttp CCATCTACGAGAGCCTCCAG CGTGGTCGGATGACAGGT
Lias AGGAAACTTAAAGCGCCAGA GCCATGGAGGTAGTCTTAGCC
Hmox1 AGGCTTTAAGCTGGTGATGG CTTCCAGGGCCGTGTAGATA
Hmox2 TGGCACCAGAAAAGGAAAAC CTTCCTTGGTCCCTTCCTTC
Alas1 TGGGGCCAAGCCAGCTCCTC AGGAGCCTGCTGGACTGCGG
Alas2 TTGGTTCGTCCTCAGTGCAGGG GCCACCATCCTGAGCCCAAAGTC
Fech CCGGAAATGCTTTCGGCCAGCG ACAGGCCCTTGAGCTGCCGTG
Alad CGCTTTTAGAGCGGGAGAGC AGAGAAGTCTGCTGGATGGAG
Pbgd AGCTACAGAGAAAGTTCCCC ACTGAATTCCTGCAGCTCAT
Uros GAGGACTCATTTTCACCAGC GACTTGGCATTCCATCTGTC
Urod TTACTGTTTACCATGGAGGC AGGAACGTGTCATTCTTCAG
Ppox ACCTAGCAAGTAAAGGGGTC CTTATAATGTGGTCAGCCTCC
Abcb10 ACCGGTGTGCGAGACCTTGGG TGGACACAGCCAGAAACCCAACTGC
Hamp1 TTTGCACGGGGAAGAAAGCA GTGGCTCTAGGCTATGTTTTGC
Bmp6 GCTTTGTGAACCTGGTGGAG GTCGTTGATGTGGGGAGAAC
Hfe CCTCCACGTTTCCAGATCCT CTCTGAGGCACCCATGAAGAG
Ftmt ATTTCCCCAGCCTGAAAGAT TCCCATCCCTGTCATACCAT
Open in a new tab

Tissues from young and aged mice

Tissues of liver, gastrocnemius muscle, and dissected brain cortex from young (4 months old) and aged (22 months old) female UM-HET3 mice, which is a genetically heterogeneous mouse model that is the first-generation offspring of a CByB6F1 × C3D2F1 cross to produce a diverse heterogeneous population in the Interventions Testing Program (ITP; https://www.nia.nih.gov/research/dab/interventions-testing-program-itp) (Harrison et al., 2021), were used in the present study. Mice were given diet Purina 5LG6 containing 345 ppm iron, which is similar to the published normal iron diet (Gutschow et al., 2015). For hepcidin immunohistological analysis, young (4 months old) and aged (22 months old) female C57BL/6 mice were used. This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. Our research was approved by the University of Michigan’s Institutional Animal Care and Use Committee. The approval number currently associated with this activity is PRO00009981.

Cytosolic and mitochondrial fractioning

Cytosolic and Mitochondrial fractions from tissues were isolated using Mitochondrial Isolation Kit for Tissues (Pierce) according to the manufacturer’s protocol.

Non-heme and heme iron measurement

Iron contents in the cytosolic and mitochondrial fractions of the tissues were measured as previously described (Bayeva et al., 2012; Sato et al., 2018). Briefly, for the non-heme iron measurements, equal amounts of protein were mixed with protein precipitation solution (1:1 1 N HCl and 10% trichloroacetic acid) and heated at 95°C for 1 hr to release iron. The precipitated proteins were removed by centrifugation at 16,000×g for 10 min at 4°C, the supernatant was mixed with the equal volume of chromogenic solution (0.5 mM ferrozine, 1.5 M sodium acetate, 0.1% [v/v] thioglycolic acid), and the absorbance was measured at 562 nm. For the heme iron measurements, equal amounts of protein were mixed with 2.0 M oxalic acid and were heated at 95°C for 1 hr to release iron from heme and generate protoporphyrin IX. Samples were then centrifuged at 1000×g for 10 minat 4°C and the fluorescence of the supernatant was assessed at 405 nm/600 nm.

Reverse transcription and quantitative real-time PCR

RNA was isolated from tissues using RNA-STAT60 (Teltest). Reverse transcribed with qScript Reverse Transcription Kit (Quanta) according to the manufacturers’ instruction, and as described previously (Chang et al., 2021). The resulting cDNA was amplified quantitatively using PerfeCTa SYBR Green Mix (Quanta) on a 7500 Fast Real-Time PCR System (Applied Biosystems). The relative gene expression was determined using differences in Ct values between gene of interest and housekeeping control genes β-actin and hypoxanthine phosphoribosyltransferase-1 (Hprt1).

Immunoblotting

Tissues were lysed in radioimmunoprecipitation assay (RIPA) buffer supplemented with protease inhibitor (Thermo Fisher Scientific). Protein concentration in lysates was determined using BCA Protein Quantification Kit (Pierce) and as described previously (Sawicki et al., 2018). Equal amounts of proteins were resolved on 4–12% Novex Bis-Tris poly-acrylamide gel (Invitrogen) and blotted onto nitrocellulose membrane (Invitrogen). After blocking with tris-buffered saline containing 0.05% Tween 20 (Thermo Fisher Scientific) and 5% milk, the membrane was incubated overnight at 4°C in primary antibody against HPRT (ProteinTech), succinate dehydrogenase complex flavoprotein subunit A (SDHA, Invitrogen), FTN (Sigma-Aldrich), IRP2 (Novus Biologicals), TfR1 (Invitrogen), FPN1 (Novus Biologicals), Hepcidin (Abcam), Ubiquitin (Cell Signaling Technology), GAPDH (ProteinTech), and α-Tubulin (ProteinTech). The following day, membranes were incubated with HRP-conjugated anti-mouse or anti-rabbit secondary antibodies (Jackson ImmunoResearch) for 1 hr and proteins were visualized using Super Surgical Western Pico ECL substrate (Pierce). Quantification of immunoblotting image was done using ImageJ (NIH).

Immunoprecipitation of FPN1

Tissues were lysed in IP buffer containing 25 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 0.1% NP-40 with protease inhibitor cocktail (G-Biosciences). 400 μg of protein was pre-incubated with 40 μl of protein G magnetic beads (Invitrogen) for 1 hr at 4°C. After the beads had been discarded, the supernatant was incubated with FPN antibody (Novus Biologicals) or IgG at a rotator overnight at 4°C. The mixture was then incubated with 40 μl of fresh beads for 1 hr at 4°C. A magnetic field was applied to this mixture, and the supernatant was removed. The magnetic beads were washed three times with 400 μl of IP buffer, re-suspended in 40 μl of SDS sample buffer, and incubated for 10 min at 70°C. Finally, 20 μl of the supernatant was collected after applying a magnetic field to the mixture and was used for immunoblotting.

Immunohistochemistry of hepcidin

Mice were anesthetized with an intraperitoneal injection of 250 mg/kg dose of freshly prepared Tribromoethanol (Avertin). They were then transcardially perfused with ice-cold phosphate-buffered saline (PBS) to wash out the blood, and following the discoloration of liver, the buffer was replaced to freshly prepared ice-cold 4% paraformaldehyde (PFA) with the assistance of circulating pump to supply sufficient perfusion pressure. After 20 min perfusion, the brain tissue was extracted from the skull and was fixed overnight in 4% PFA. The fixed brain tissue was placed in 30% sucrose for 48 hr at 4°C. Brain tissue was then submerged into OTC compound (Sakura Finetek, Torrance, CA) and frozen in liquid nitrogen. 30 µm coronal sections were cut using a cryotome (Leica) and mounted on Fisherbrand Superfrost Plus Microscope Slides. Sections were permeabilized using 0.25% Triton-x100 in PBS for 30 min, washed three times in PBS, and incubated with primary antibody against hepcidin overnight at 4°C. Sections were then washed three times in PBS and incubated with Alexa Fluor594 Goat Anti-Rabbit IgG at 1:200 (Jackson ImmunoResearch) for 2 hr at room temperature. Nuclei were counterstained with DAPI containing ProLong Gold Antifade mounting media (Invitrogen) and images of frontal cortex were acquired using a Zeiss Axio Observer.Z1 fluorescence microscope.

Aconitase activity measurements

Aconitase activities in cytosolic and mitochondrial fraction of the brain cortex lysates were measured using Aconitase Enzyme Activity Microplate Assay Kit (Abcam) according to the manufacturer’s instructions.

Statistical analysis

Data are presented as mean ± SEM. There was no randomization of the samples, but the investigator performing the measurements was blinded to the group assignment. For sample size estimation, our previous study (Chang et al., 2016) indicated that changes in mitochondrial non-heme iron contents of about 20% can affect iron-dependent cytotoxicity in the mouse heart. Given an estimated standard deviation of iron of 15% in each group, we estimated that six samples would be needed to detect a 20% difference in change of iron contents. For other experiments, power analysis was not performed a priori because we could use sufficient numbers of stored tissue samples that had already been cryopreserved and there was no risk of wasting unnecessary animals. We determined n=10 for the RNA levels and n=3–6 for the immunoblots and enzyme activities for our analysis. This was based on our prior experience performing similar experiments. Unpaired two-tailed Student’s t-tests were used to determine statistical significance. p<0.05 was considered to be statistically significant, as indicated by an asterisk. Analysis was performed using Graphpad Prism 9.

Acknowledgements

The authors would like to thank Chunlei Chen for technical help. TS was supported by American Heart Association. HA is supported by NIH R01 HL127646, R01 HL140973, and R01 HL138982, and a grant from Leducq foundation. RAM was supported by NIH grants U01-AG022303-17 and P30-AG024824.

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Contributor Information

Hossein Ardehali, Email: h-ardehali@northwestern.edu.

Yelena Ginzburg, Icahn School of Medicine at Mount Sinai, United States.

Mone Zaidi, Icahn School of Medicine at Mount Sinai, United States.

Funding Information

This paper was supported by the following grants:

  • NIH Office of the Director NIH R01 HL127646 to Hossein Ardehali.

  • Leducq Foundation to Hossein Ardehali.

  • NIH Office of the Director R01 HL140973 to Hossein Ardehali.

  • NIH Office of the Director R01 HL138982 to Hossein Ardehali.

Additional information

Competing interests

Reviewing editor, eLife.

No competing interests declared.

Author contributions

Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing - review and editing.

Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing - review and editing.

Conceptualization, Methodology, Validation, Writing - review and editing.

Formal analysis, Resources, Writing - review and editing.

Conceptualization, Formal analysis, Funding acquisition, Investigation, Project administration, Supervision, Validation, Writing - original draft, Writing - review and editing.

Ethics

This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All of the animals were handled according to approved institutional animal care and use committee (IACUC) of Northwestern University. PRO00009981.

Additional files

Transparent reporting form

Data availability

All data generated or analysed during this study were included as figures in the paper. All of the original datasets are included in the manuscript and supporting files.

References

  1. Bartzokis G, Mintz J, Sultzer D, Marx P, Herzberg JS, Phelan CK, Marder SR. In vivo MR evaluation of age-related increases in brain iron. AJNR. American Journal of Neuroradiology. 1994a;15:1129–1138. [PMC free article] [PubMed] [Google Scholar]
  2. Bartzokis G., Sultzer D, Mintz J, Holt LE, Marx P, Phelan CK, Marder SR. In vivo evaluation of brain iron in Alzheimer’s disease and normal subjects using MRI. Biological Psychiatry. 1994b;35:480–487. doi: 10.1016/0006-3223(94)90047-7. [DOI] [PubMed] [Google Scholar]
  3. Bartzokis G, Lu PH, Tishler TA, Peters DG, Kosenko A, Barrall KA, Finn JP, Villablanca P, Laub G, Altshuler LL, Geschwind DH, Mintz J, Neely E, Connor JR. Prevalent iron metabolism gene variants associated with increased brain ferritin iron in healthy older men. Journal of Alzheimer’s Disease. 2010;20:333–341. doi: 10.3233/JAD-2010-1368. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bayeva M, Khechaduri A, Puig S, Chang HC, Patial S, Blackshear PJ, Ardehali H. mTOR regulates cellular iron homeostasis through tristetraprolin. Cell Metabolism. 2012;16:645–657. doi: 10.1016/j.cmet.2012.10.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Benkovic SA, Connor JR. Ferritin, transferrin, and iron in selected regions of the adult and aged rat brain. The Journal of Comparative Neurology. 1993;338:97–113. doi: 10.1002/cne.903380108. [DOI] [PubMed] [Google Scholar]
  6. Berg D, Gerlach M, Youdim MB, Double KL, Zecca L, Riederer P, Becker G. Brain iron pathways and their relevance to Parkinson’s disease. Journal of Neurochemistry. 2001;79:225–236. doi: 10.1046/j.1471-4159.2001.00608.x. [DOI] [PubMed] [Google Scholar]
  7. Chang H-C, Wu R, Shang M, Sato T, Chen C, Shapiro JS, Liu T, Thakur A, Sawicki KT, Prasad SVN, Ardehali H. Reduction in mitochondrial iron alleviates cardiac damage during injury. EMBO Molecular Medicine. 2016;8:247–267. doi: 10.15252/emmm.201505748. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chang HC, Shapiro JS, Jiang X, Senyei G, Sato T, Geier J, Sawicki KT, Ardehali H. Augmenter of liver regeneration regulates cellular iron homeostasis by modulating mitochondrial transport of ATP-binding cassette B8. eLife. 2021;10:e65158. doi: 10.7554/eLife.65158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chaudhary S, Ashok A, McDonald D, Wise AS, Kritikos AE, Rana NA, Harding CV, Singh N. Upregulation of Local Hepcidin Contributes to Iron Accumulation in Alzheimer’s Disease Brains. Journal of Alzheimer’s Disease. 2021;82:1487–1497. doi: 10.3233/JAD-210221. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Flurkey K, Currer JM, Harrison DE. In: ” In The Mouse in Biomedical Research, Eds. Davisson MT, Quimby FW, Barthold SW, Newcomer CE, Smith AL, editors. Amsterdam: Elsevier Inc; 2007. Mouse models in aging research; pp. 637–672. [DOI] [Google Scholar]
  11. Gutschow P, Schmidt PJ, Han H, Ostland V, Bartnikas TB, Pettiglio MA, Herrera C, Butler JS, Nemeth E, Ganz T, Fleming MD, Westerman M. A competitive enzyme-linked immunosorbent assay specific for murine hepcidin-1: correlation with hepatic mRNA expression in established and novel models of dysregulated iron homeostasis. Haematologica. 2015;100:167–177. doi: 10.3324/haematol.2014.116723. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hahn P, Song Y, Ying G, He X, Beard J, Dunaief JL. Age-dependent and gender-specific changes in mouse tissue iron by strain. Experimental Gerontology. 2009;44:594–600. doi: 10.1016/j.exger.2009.06.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Harrison DE, Strong R, Reifsnyder P, Kumar N, Fernandez E, Flurkey K, Javors MA, Lopez-Cruzan M, Macchiarini F, Nelson JF, Bitto A, Sindler AL, Cortopassi G, Kavanagh K, Leng L, Bucala R, Rosenthal N, Salmon A, Stearns TM, Bogue M, Miller RA. 17-a-estradiol late in life extends lifespan in aging UM-HET3 male mice; nicotinamide riboside and three other drugs do not affect lifespan in either sex. Aging Cell. 2021;20:e13328. doi: 10.1111/acel.13328. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kandlur A, Satyamoorthy K, Gangadharan G. Oxidative Stress in Cognitive and Epigenetic Aging: A Retrospective Glance. Frontiers in Molecular Neuroscience. 2020;13:41. doi: 10.3389/fnmol.2020.00041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Kawabata H, Yang R, Hirama T, Vuong PT, Kawano S, Gombart AF, Koeffler HP. Molecular cloning of transferrin receptor 2. A new member of the transferrin receptor-like family. The Journal of Biological Chemistry. 1999;274:20826–20832. doi: 10.1074/jbc.274.30.20826. [DOI] [PubMed] [Google Scholar]
  16. Koleini N, Shapiro JS, Geier J, Ardehali H. Ironing out mechanisms of iron homeostasis and disorders of iron deficiency. The Journal of Clinical Investigation. 2021;131:148671. doi: 10.1172/JCI148671. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lakhal-Littleton S, Wolna M, Chung YJ, Christian HC, Heather LC, Brescia M, Ball V, Diaz R, Santos A, Biggs D, Clarke K, Davies B, Robbins PA. An essential cell-autonomous role for hepcidin in cardiac iron homeostasis. eLife. 2016;5:e19804. doi: 10.7554/eLife.19804. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Lei P, Ayton S, Finkelstein DI, Spoerri L, Ciccotosto GD, Wright DK, Wong BXW, Adlard PA, Cherny RA, Lam LQ, Roberts BR, Volitakis I, Egan GF, McLean CA, Cappai R, Duce JA, Bush AI. Tau deficiency induces parkinsonism with dementia by impairing APP-mediated iron export. Nature Medicine. 2012;18:291–295. doi: 10.1038/nm.2613. [DOI] [PubMed] [Google Scholar]
  19. Liang T, Qian ZM, Mu MD, Yung WH, Ke Y. Brain Hepcidin Suppresses Major Pathologies in Experimental Parkinsonism. IScience. 2020;23:101284. doi: 10.1016/j.isci.2020.101284. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lieblein-Boff JC, McKim DB, Shea DT, Wei P, Deng Z, Sawicki C, Quan N, Bilbo SD, Bailey MT, McTigue DM, Godbout JP. Neonatal E. coli infection causes neuro-behavioral deficits associated with hypomyelination and neuronal sequestration of iron. The Journal of Neuroscience. 2013;33:16334–16345. doi: 10.1523/JNEUROSCI.0708-13.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lu LN, Qian ZM, Wu KC, Yung WH, Ke Y. Expression of Iron Transporters and Pathological Hallmarks of Parkinson’s and Alzheimer’s Diseases in the Brain of Young, Adult, and Aged Rats. Molecular Neurobiology. 2017;54:5213–5224. doi: 10.1007/s12035-016-0067-0. [DOI] [PubMed] [Google Scholar]
  22. Mastroberardino PG, Hoffman EK, Horowitz MP, Betarbet R, Taylor G, Cheng D, Na HM, Gutekunst C-A, Gearing M, Trojanowski JQ, Anderson M, Chu CT, Peng J, Greenamyre JT. A novel transferrin/TfR2-mediated mitochondrial iron transport system is disrupted in Parkinson’s disease. Neurobiology of Disease. 2009;34:417–431. doi: 10.1016/j.nbd.2009.02.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Noster J, Persicke M, Chao TC, Krone L, Heppner B, Hensel M, Hansmeier N. Impact of ROS-Induced Damage of TCA Cycle Enzymes on Metabolism and Virulence of Salmonella enterica serovar Typhimurium. Frontiers in Microbiology. 2019;10:762. doi: 10.3389/fmicb.2019.00762. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Nyberg L, Salami A, Andersson M, Eriksson J, Kalpouzos G, Kauppi K, Lind J, Pudas S, Persson J, Nilsson LG. Longitudinal evidence for diminished frontal cortex function in aging. PNAS. 2010;107:22682–22686. doi: 10.1073/pnas.1012651108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Pellegrino RM, Boda E, Montarolo F, Boero M, Mezzanotte M, Saglio G, Buffo A, Roetto A. Transferrin Receptor 2 Dependent Alterations of Brain Iron Metabolism Affect Anxiety Circuits in the Mouse. Scientific Reports. 2016;6:30725. doi: 10.1038/srep30725. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Qian ZM, Wang Q. Expression of iron transport proteins and excessive iron accumulation in the brain in neurodegenerative disorders. Brain Research. Brain Research Reviews. 1998;27:257–267. doi: 10.1016/s0165-0173(98)00012-5. [DOI] [PubMed] [Google Scholar]
  27. Qiao B, Sugianto P, Fung E, Del-Castillo-Rueda A, Moran-Jimenez MJ, Ganz T, Nemeth E. Hepcidin-induced endocytosis of ferroportin is dependent on ferroportin ubiquitination. Cell Metabolism. 2012;15:918–924. doi: 10.1016/j.cmet.2012.03.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Roskams AJ, Connor JR. Iron, transferrin, and ferritin in the rat brain during development and aging. Journal of Neurochemistry. 1994;63:709–716. doi: 10.1046/j.1471-4159.1994.63020709.x. [DOI] [PubMed] [Google Scholar]
  29. Samavati L, Lee I, Mathes I, Lottspeich F, Hüttemann M. Tumor necrosis factor alpha inhibits oxidative phosphorylation through tyrosine phosphorylation at subunit I of cytochrome c oxidase. The Journal of Biological Chemistry. 2008;283:21134–21144. doi: 10.1074/jbc.M801954200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Sato T, Chang H-C, Bayeva M, Shapiro JS, Ramos-Alonso L, Kouzu H, Jiang X, Liu T, Yar S, Sawicki KT, Chen C, Martínez-Pastor MT, Stumpo DJ, Schumacker PT, Blackshear PJ, Ben-Sahra I, Puig S, Ardehali H. mRNA-binding protein tristetraprolin is essential for cardiac response to iron deficiency by regulating mitochondrial function. PNAS. 2018;115:E6291–E6300. doi: 10.1073/pnas.1804701115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Sawicki KT, Chang HC, Shapiro JS, Bayeva M, De Jesus A, Finck BN, Wertheim JA, Blackshear PJ, Ardehali H. Hepatic tristetraprolin promotes insulin resistance through RNA destabilization of FGF21. JCI Insight. 2018;3:95948. doi: 10.1172/jci.insight.95948. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Schwartz AJ, Goyert JW, Solanki S, Kerk SA, Chen B, Castillo C, Hsu PP, Do BT, Singhal R, Dame MK, Lee H-J, Spence JR, Lakhal-Littleton S, Heiden MGV, Lyssiotis CA, Xue X, Shah YM. Hepcidin sequesters iron to sustain nucleotide metabolism and mitochondrial function in colorectal cancer epithelial cells. Nature Metabolism. 2021;3:969–982. doi: 10.1038/s42255-021-00406-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Silvestri L, Nai A, Pagani A, Camaschella C. The extrahepatic role of TFR2 in iron homeostasis. Frontiers in Pharmacology. 2014;5:93. doi: 10.3389/fphar.2014.00093. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Smith MA, Harris PL, Sayre LM, Perry G. Iron accumulation in Alzheimer disease is a source of redox-generated free radicals. PNAS. 1997;94:9866–9868. doi: 10.1073/pnas.94.18.9866. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Sun Y, Pham AN, Waite TD. Mechanism Underlying the Effectiveness of Deferiprone in Alleviating Parkinson’s Disease Symptoms. ACS Chemical Neuroscience. 2018;9:1118–1127. doi: 10.1021/acschemneuro.7b00478. [DOI] [PubMed] [Google Scholar]
  36. Wang J, Chen G, Muckenthaler M, Galy B, Hentze MW, Pantopoulos K. Iron-mediated degradation of IRP2, an unexpected pathway involving a 2-oxoglutarate-dependent oxygenase activity. Molecular and Cellular Biology. 2004;24:954–965. doi: 10.1128/MCB.24.3.954-965.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Wang Q, Du F, Qian ZM, Ge XH, Zhu L, Yung WH, Yang L, Ke Y. Lipopolysaccharide induces a significant increase in expression of iron regulatory hormone hepcidin in the cortex and substantia nigra in rat brain. Endocrinology. 2008;149:3920–3925. doi: 10.1210/en.2007-1626. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Xiong XY, Liu L, Wang FX, Yang YR, Hao JW, Wang PF, Zhong Q, Zhou K, Xiong A, Zhu WY, Zhao T, Meng ZY, Wang YC, Gong QW, Liao MF, Wang J, Yang QW. Toll-Like Receptor 4/MyD88-Mediated Signaling of Hepcidin Expression Causing Brain Iron Accumulation, Oxidative Injury, and Cognitive Impairment After Intracerebral Hemorrhage. Circulation. 2016;134:1025–1038. doi: 10.1161/CIRCULATIONAHA.116.021881. [DOI] [PubMed] [Google Scholar]
  39. Xu J, Jia Z, Knutson MD, Leeuwenburgh C. Impaired iron status in aging research. International Journal of Molecular Sciences. 2012;13:2368–2386. doi: 10.3390/ijms13022368. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Xu J, Xiao C, Song W, Cui X, Pan M, Wang Q, Feng Y, Xu Y. Elevated Heme Oxygenase-1 Correlates With Increased Brain Iron Deposition Measured by Quantitative Susceptibility Mapping and Decreased Hemoglobin in Patients With Parkinson’s Disease. Frontiers in Aging Neuroscience. 2021;13:656626. doi: 10.3389/fnagi.2021.656626. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zecca L, Gallorini M, Schünemann V, Trautwein AX, Gerlach M, Riederer P, Vezzoni P, Tampellini D. Iron, neuromelanin and ferritin content in the substantia nigra of normal subjects at different ages: consequences for iron storage and neurodegenerative processes. Journal of Neurochemistry. 2001;76:1766–1773. doi: 10.1046/j.1471-4159.2001.00186.x. [DOI] [PubMed] [Google Scholar]
  42. Zhang X, Gou Y-J, Zhang Y, Li J, Han K, Xu Y, Li H, You L-H, Yu P, Chang Y-Z, Gao G. Hepcidin overexpression in astrocytes alters brain iron metabolism and protects against amyloid-β induced brain damage in mice. Cell Death Discovery. 2020;6:113. doi: 10.1038/s41420-020-00346-3. [DOI] [PMC free article] [PubMed] [Google Scholar]

Editor's evaluation

Yelena Ginzburg 1

The authors present a manuscript addressing an important unmet need, aiming to understand the mechanism of neurodegenerative disease, specifically focused on the role of iron overload in the aging brain, that is of interest to basic iron biologists and neurologists as a Short Report of novel significance. The current study demonstrates increased cytosolic and mitochondrial non-heme iron only in the aging brain, increased cortical hepcidin expression, and decreased levels of FPN1, together supporting a hypothesis that cortical hepcidin sequesters iron in neuronal cells and is associated with aging.

Decision letter

Editor: Yelena Ginzburg1
Reviewed by: Joshua L Dunaief2

Our editorial process produces two outputs: i) public reviews designed to be posted alongside the preprint for the benefit of readers; ii) feedback on the manuscript for the authors, including requests for revisions, shown below. We also include an acceptance summary that explains what the editors found interesting or important about the work.

Decision letter after peer review:

Thank you for submitting your article "Aging is associated with increased brain iron through brain- derived hepcidin expression" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by Mone Zaidi as the Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Joshua L Dunaief (Reviewer #2).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

The reviewers found significant merit in the work. However, the evaluation identified sufficient weaknesses that prevent publication of this work in the present form. The main shortcomings of the presented work include the relatively underdeveloped cell-specificity of hepcidin expression in the brain and lack of mechanistic information explaining hepcidin upregulation. In this Short Report format, it would be reasonable for the authors to extend the discussion regarding the shortcomings of the current approach, especially addressing the prior art and aligning a more modest conclusion given the relatively superficial presentation. In addition, it would be essential that the revision included: (1) delineating which part of the brain was dissected for these studies, (2) validation of the purity / enrichment of the mitochondrial fraction, (3) addition of means or changing data to δ δ Ct in place of "mRNA expression" in Figure 2A and B graphs, (4) discussion of why TFR1 is not decreased in the aged iron overloaded mice (Figure 3A), and (5) including additional corroborating markers of oxidative stress (in addition to aconitase activity) would increase enthusiasm. A complete list of reviewer comments can be found below.

Reviewer #1 (Recommendations for the authors):

Why is it important that cytosolic heme iron in the muscle is not found? Can the authors clarify and add to the Discussion section?

Decreased Steap3, an endosomal ferrireductase, is associated with anaemia due to decreased import of ion into erythroblasts (Blanc AJH 2015). This is worth at least a mention in the discussion.

The authors state that the same genes in the liver are not related to iron loading as the genes in the brain that are (e.g. Hox1 and Alad). Please edit.

As changes in TFR2 are possible evidence of a separate topic, this reviewer would recommend perhaps changing the language in this part of the Results section to state that TFR2 is being fully explored in a separate manuscript or the like.

Reviewer #2 (Recommendations for the authors):

1) Cite a journal article instead of: (https://www.jax.org/news-and-insights/jaxblog/2017/november/when-are-mice-considered-old#).

2) The following statement is necessarily correct: "These results suggest that the more oxidative form of iron (i.e., non-heme iron) is increased in the brain cytosol and mitochondria with aging." Heme can also be an oxidant.

3) In one place, ferroportin is misspelled as "feroportin"

4) Means should be shown on the graph for Figure 2 A and B

Reviewer #3 (Recommendations for the authors):

The manuscript by Sato et al. addresses an important question related to brain iron accumulation with aging. This is a timely question because iron accumulates in the brains of aged individuals and is believed to contribute to neurodegenerative diseases such as AD and PD.

However, the data are not clear, and do not support the major conclusion that local hepcidin is responsible for iron accumulation in the brain. Some examples are given below:

1. In Figure 1, the ferrozine method used to quantitate iron levels also measures iron released from ferritin. Iron sequestered in ferritin is in the ferric form, and not redox-active. Levels of ferritin in the brain increase with ageing, which is well-known, and could contribute to the observed increase in iron with aging noted by the authors.

2. The mitochondrial fraction was not tested for purity. Mitochondrial markers should be evaluated and compared with the cytosolic fraction to determine the efficacy of the procedure.

3. In Figure 2 A and B, the fold change in mRNA expression is not as accurate as qPCR analysis. In the methods section the authors mention that qPCR was used. In that case, the results should be presented as ΔΔCt instead of mRNA expression. Moreover, the change in mRNA expression of HAMP1 is marginal given the standard deviation.

4. In Figure 2 C, the hepcidin band is not clear. In the raw data for Figure 2, it seems that recombinant hepcidin peptide was fractionated in parallel in the last lane, though it is not labeled. It is unclear why the bands marked in the last lane of the 'middle exposure' are absent in the overnight longer exposure.

5. The brain homogenate should be depleted of capillary endothelial cells that synthesize hepcidin, or appropriate controls need to be used to remove their contribution.

6. In the immunostained brain sample in Figure 2 E, a positive control, perhaps a section of the liver from iron depleted and iron overloaded mouse would provide confidence that the reaction is from hepcidin, though it could be hepcidin from the peripheral circulation.

7. In Figure 3 A, TfR1 is expected to decrease if the cells are iron loaded. The authors need to discuss this discrepancy.

In conclusion, though the study is well conceived and asks an important question, the experimental design lacks appropriate controls, and the results do not support the conclusions drawn by the authors.

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

Thank you for submitting your article "Aging is associated with increased brain iron through brain- derived hepcidin expression" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, and the evaluation has been overseen by a Reviewing Editor and Mone Zaidi as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Joshua L Dunaief (Reviewer #2).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

The reviewers appreciate the much improved revision of this manuscript. Several additional components need to be addressed.

1) Findings of significantly decreased FPN in the western from this outbred strain with N=3 biological replicates need to be repeated and confirmed. In lieu of a repeated experiment, the authors may choose to add to the Discussion section on this point that "additional experiments are needed to confirm these findings in a larger cohort of mice of both genders."

2) Given the authors' response regarding the selection of the frontal cortex as the main site of analysis, the Introduction is no longer relevant and should be edited to fall in line with the current focus, not focused on the substantia nigra and striatum.

Reviewer #1 (Recommendations for the authors):

The authors have substantively responded to most of the reviewers' comments. Several elements remain to be resolved.

1) The authors suggest in their response that they focused on the frontal cortex due to prior evidence of the importance of cortical atrophy with age. However, the introduction has a very different read and in fact states that the "substantia nigra and stratum" (misspelled striatum?), subcortical structures, are central to disease pathophysiology. Please edit the introduction more in line with the selected area of the brain studied in the current manuscript and to remove the substantive delineation of Alzheimer's and Parkinson's diseases which are NOT the focus of the current manuscript.

2) Change the title to "Aging is associated with increased brain iron through cortex-derived hepcidin expression."

3) Please spell out HPRT1 and SDCF in the methods section.

4) Discussion section regarding TFR1 regulation: please change "HIF-medicated" to "HIF-mediated".

5) Please add the following sentence to the Discussion section regarding TFR1, at the end: "Our data therefore cannot resolve whether TFR1 is central to age-mediated iron accumulation in the brain and if so, by what mechanism. Additional studies are necessary to clarify this important point." Alternatively, please measure other pathways that regulate TFR1.

6) Aconitase is involved in iron responsiveness regulation in other tissues (Bullock Blood 2010). As a consequence, using aconitase as a read out of oxidative stress when iron is altered between the 2 groups is insufficient. Please remove the phrase "indicating increased oxidative stress" in the last sentence of the Results section.

7) Please add "Further studies… increased TFR2 in the aged brain contributes to increased brain-derived hepcidin expression or alternatively contributes to iron accumulation with age." to the TFR2 part of the Discussion section.

8) Question about muscle was intended to understand how data in muscle is relevant in a paper about the brain. This remains unclear.

9) Please add components of mouse strain and gender and haematologica reference for iron diet to the methods section.

Reviewer #2 (Recommendations for the authors):

The authors have addressed a number of concerns, including identifying the region of brain studied and validating the enriched mitochondrial fraction. Identifying the cell types with increased Hamp1/Hepc mRNA and protein and decreased Fpn protein is important, but this has not been done. In addition, the important finding of decreased Fpn is shown in a Western with an N of only 3. This is inadequate, especially with an outbred strain and documented high variability in Hamp1/Hepc levels in Figure 2. The N for the Fpn Western should be increased, and should be correlated with Hamp1/Hepc levels.

eLife. 2022 Jan 11;11:e73456. doi: 10.7554/eLife.73456.sa2

Author response

Essential revisions:

The reviewers found significant merit in the work. However, the evaluation identified sufficient weaknesses that prevent publication of this work in the present form. The main shortcomings of the presented work include the relatively underdeveloped cell-specificity of hepcidin expression in the brain and lack of mechanistic information explaining hepcidin upregulation. In this Short Report format, it would be reasonable for the authors to extend the discussion regarding the shortcomings of the current approach, especially addressing the prior art and aligning a more modest conclusion given the relatively superficial presentation. In addition, it would be essential that the revision included:

1) Delineating which part of the brain was dissected for these studies,

Thank you for this comment. In the present study, we used frozen tissue from brain cortex for iron measurement and frontal cortex for immunohistochemistry of hepcidin. The purpose of the present study was to explore if and how cellular iron levels change in the brain with aging, but not in the brain of neurodegenerative diseases such as Alzheimer's or Parkinson's. Since it has been reported that age-related brain atrophy (which is associated with age-related cognitive decline) is evident in the cortex (specifically in the frontal cortex) and not in other parts of the brain (Proc Natl Acad Sci U S A. 2010 Dec 28;107(52):22682-6), we focused our studies on the frontal cortex. Our studies revealed that increased hepcidin is associated with iron-accumulation in the aged brain cortex. The information about the part of the brain is now added to the manuscript.

2) Validation of the purity / enrichment of the mitochondrial fraction,

We appreciate the careful review of our data. We have performed immunoblotting experiments using cytosolic marker (HPRT1 = hypoxanthine phosphoribosyltransferase-1) and mitochondrial marker (Succinate Dehydrogenase Complex Flavoprotein Subunit A) in the samples used for iron measurement, and confirmed the purity of mitochondrial isolation as shown in the revised Figure 1G.

3) Addition of means or changing data to δ δ Ct in place of "mRNA expression" in Figure 2A and B graphs,

Thank you for this comment. We have added the “means and SEM” in the revised Figure 2A and 2B.

4) Discussion of why TFR1 is not decreased in the aged iron overloaded mice (Figure 3A).

It has been known that TfR1 is regulated by both iron-dependent and iron-independent mechanisms. For iron-dependent pathways, IRPs are well-known positive regulators of TfR1. Since our data showed that the protein level of IRP2 was decreased in the aged brain, it is plausible that TfR1 would be concomitantly downregulated via decreased IRP2 activity. However, TfR1 has also been known to be regulated by iron-independent immune response, such as NF-κB- and/or HIF-medicated pathways (J Biol Chem. 2008 Jul 25;283(30):20674-86). Thus, it is likely that activated iron-independent regulation of TfR1 cancels IRP2-medicated regulation of TfR1, resulting in unchanged TfR1 expression in the aged brain. This discussion is now added to the revised manuscript.

5) Including additional corroborating markers of oxidative stress (in addition to aconitase activity) would increase enthusiasm. A complete list of reviewer comments can be found below.

We appreciate this suggestion. It has already been established that oxidative stress is increased in the aged brain, as summarized in the following review (Front Mol Neurosci. 2020 Mar 18;13:41). In addition, our data already suggest the presence of oxidative stress in the aged brain by measuring both cytoplasmic and mitochondrial aconitase activities. Therefore, we did not measure canonical oxidative stress markers such as lipid-peroxidation in the present study, and instead, cite previous reports that show age-related oxidative stress, in addition to the inclusion of aconitase activity.

Reviewer #1 (Recommendations for the authors):

Why is it important that cytosolic heme iron in the muscle is not found? Can the authors clarify and add to the Discussion section?

The precise reason why we were not able to detect heme iron in the mitochondria of gastrocnemius muscle is unclear, but it could possibly due to technical limitations associated with the isolation of intact mitochondria from skeletal muscle tissue. Since gastrocnemius muscle is a fast-twitch muscle containing low concentration of mitochondria, it is also possible that low mitochondrial yield was insufficient for detecting measurable iron, especially heme iron.

Decreased Steap3, an endosomal ferrireductase, is associated with anaemia due to decreased import of ion into erythroblasts (Blanc AJH 2015). This is worth at least a mention in the discussion.

Thank you very much for this suggestion. We hope that the Reviewer agrees that since our paper focuses on brain iron and Steap3 is not changed in the brain with aging, we do not think a discussion on this protein would be relevant to the paper.

The authors state that the same genes in the liver are not related to iron loading as the genes in the brain that are (e.g. Hox1 and Alad). Please edit.

In the revised manuscript, we discuss the possible roles of HOX1 and ALAD in iron and aging.

As changes in TFR2 are possible evidence of a separate topic, this reviewer would recommend perhaps changing the language in this part of the Results section to state that TFR2 is being fully explored in a separate manuscript or the like.

Thank you for this great suggestion. In the revised manuscript, we have mentioned possible roles of TfR2 in iron accumulation in aged brain citing previous papers.

Reviewer #2 (Recommendations for the authors):

1) Cite a journal article instead of: (https://www.jax.org/news-and-insights/jaxblog/

2017/november/when-are-mice-considered-old#).

We thank the reviewer for this suggestion. In the revised manuscript, we cited original textbook from “Flurkey K, Currer JM, Harrison DE. 2007. The Mouse in Aging Research. In The Mouse in Biomedical Research 2nd Edition. Fox JG, et al., editors. American College Laboratory Animal Medicine (Elsevier), Burlington, MA. pp. 637–672.”, instead of citing URL.

2) The following statement is necessarily correct: "These results suggest that the more oxidative form of iron (i.e., non-heme iron) is increased in the brain cytosol and mitochondria with aging." Heme can also be an oxidant.

We agree with the Reviewer that heme iron can also exert oxidative stress, and this statement is revised in the manuscript.

3) In one place, ferroportin is misspelled as "feroportin".

We apologize for the mistake. In the revised manuscript, this has been corrected.

4) Means should be shown on the graph for Figure 2 A and B.

As we responded in the “Essential Revision-3”, means and SEM have been included in the revised Figure 2A and B.

Reviewer #3 (Recommendations for the authors):

The manuscript by Sato et al. addresses an important question related to brain iron accumulation with aging. This is a timely question because iron accumulates in the brains of aged individuals and is believed to contribute to neurodegenerative diseases such as AD and PD.

However, the data are not clear, and do not support the major conclusion that local hepcidin is responsible for iron accumulation in the brain. Some examples are given below:

1. In Figure 1, the ferrozine method used to quantitate iron levels also measures iron released from ferritin. Iron sequestered in ferritin is in the ferric form, and not redox-active. Levels of ferritin in the brain increase with ageing, which is well-known, and could contribute to the observed increase in iron with aging noted by the authors.

Thank you for this comment. As the reviewer mentioned, our iron measurements may include ferritin-binding iron. However, our data showed that in addition to the cytoplasmic iron, mitochondrial iron (which is directly associated with cellular damage) is increased in the brain with aging, suggesting elevated total cellular iron in the aged brain. In addition, the activity of aconitase in both cytosol and mitochondria was decreased, suggesting the involvement of iron-mediated oxidative stress in the aged brain.

2. The mitochondrial fraction was not tested for purity. Mitochondrial markers should be evaluated and compared with the cytosolic fraction to determine the efficacy of the procedure.

As mentioned in our response to the “Essential Revision-2” and the reviewer’s comment of #2-additional limitation-3, we have confirmed the purity of mitochondrial isolation as shown in the revised Figure 1G.

3. In Figure 2 A and B, the fold change in mRNA expression is not as accurate as qPCR analysis. In the methods section the authors mention that qPCR was used. In that case, the results should be presented as ΔΔCt instead of mRNA expression. Moreover, the change in mRNA expression of HAMP1 is marginal given the standard deviation.

Since we assessed many genes associated with cellular iron regulation, we believe that showing the relative change is a better way to demonstrate the differences in their expression. We also provide the original data including CT values in the supplementary tables that readers can access. We acknowledge that the mRNA level of hepcidin in the aged brain, but not in the aged liver, showed increased SD, but the difference was statistically significant. We also confirmed that hepcidin protein expression is increased in Western blot and immunohistochemistry experiments.

4. In Figure 2 C, the hepcidin band is not clear. In the raw data for Figure 2, it seems that recombinant hepcidin peptide was fractionated in parallel in the last lane, though it is not labeled. It is unclear why the bands marked in the last lane of the 'middle exposure' are absent in the overnight longer exposure.

We appreciate the careful review of our data. The last lane was just a ladder and marked by a pen (3-, 6- and 10-kDa). The second lane form the right was lysate from the liver tissue, which is the major organ for hepcidin production.

5. The brain homogenate should be depleted of capillary endothelial cells that synthesize hepcidin, or appropriate controls need to be used to remove their contribution.

Thank you for this comment. As mentioned in our response to comments of #1-3 and #2-1, we acknowledge the limitation that we cannot reveal cell-specificity of hepcidin expression in the aged brain. This limitation has been included in the revised manuscript.

6. In the immunostained brain sample in Figure 2 E, a positive control, perhaps a section of the liver from iron depleted and iron overloaded mouse would provide confidence that the reaction is from hepcidin, though it could be hepcidin from the peripheral circulation.

We appreciate your careful review of our data. Our data showed that protein and mRNA levels of hepcidin are increased in the brain of aged mice compared to young mice, suggesting that brainderived hepcidin is actually up-regulated with aging. Therefore, the possibility that contamination of circulating hepcidin affected our analysis would be quite low. We hope the Reviewer agrees with no additional experiments, and that our current data are sufficient to show increased hepcidin expression in the aged brain.

7. In Figure 3 A, TfR1 is expected to decrease if the cells are iron loaded. The authors need to discuss this discrepancy.

Thank you for this comment. As we responded in the “Essential Revision-4”, our possible explanation for the unchanged TfR1 levels in spite of iron accumulation in the aged brain has been added to the Discussion section of the revised manuscript.

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

Essential revisions:

The reviewers appreciate the much improved revision of this manuscript. Several additional components need to be addressed.

1) Findings of significantly decreased FPN in the western from this outbred strain with N=3 biological replicates need to be repeated and confirmed. In lieu of a repeated experiment, the authors may choose to add to the Discussion section on this point that "additional experiments are needed to confirm these findings in a larger cohort of mice of both genders."

Thank you for this great suggestion. We have added the statement of "Since the results of this study are based on a small number of immunoblot samples using single-gender of mice, additional studies are needed to confirm these findings in a larger cohort of mice of both genders" at the end of 2nd paragraph of the Discussion section in the revised manuscript.

2) Given the authors' response regarding the selection of the frontal cortex as the main site of analysis, the Introduction is no longer relevant and should be edited to fall in line with the current focus, not focused on the substantia nigra and striatum.

We appreciate this valuable suggestion. According to the suggestion, we edited the introduction section. Especially, we removed the sentences mentioning Alzheimer's and Parkinson's diseases and substantia nigra and striatum and added the sentence mentioning the reason why we focused on the brain cortex in the present study. As a result, the four paragraphs in the previous introduction section have been combined into three in the revised manuscript.

Reviewer #1 (Recommendations for the authors):

The authors have substantively responded to most of the reviewers' comments. Several elements remain to be resolved.

1) The authors suggest in their response that they focused on the frontal cortex due to prior evidence of the importance of cortical atrophy with age. However, the introduction has a very different read and in fact states that the "substantia nigra and stratum" (misspelled striatum?), subcortical structures, are central to disease pathophysiology. Please edit the introduction more in line with the selected area of the brain studied in the current manuscript and to remove the substantive delineation of Alzheimer's and Parkinson's diseases which are NOT the focus of the current manuscript.

Thank you for this valuable suggestion. We have revised the introduction section by removing the sentences mentioning Alzheimer's and Parkinson's diseases and substantia nigra and striatum and added the sentence mentioning the reason why we focused on the brain cortex in the present study.

2) Change the title to "Aging is associated with increased brain iron through cortex-derived hepcidin expression."

We agree with the reviewer’s suggestion and the title has been changed in the revised manuscript.

3) Please spell out HPRT1 and SDCF in the methods section.

We thank for the comment, and now this is amended.

4) Discussion section regarding TFR1 regulation: please change "HIF-medicated" to "HIF-mediated".

We apologize for this typographical errors. All of "medicated" words have also been changed to “mediated” in the revised version.

5) Please add the following sentence to the Discussion section regarding TFR1, at the end: "Our data therefore cannot resolve whether TFR1 is central to age-mediated iron accumulation in the brain and if so, by what mechanism. Additional studies are necessary to clarify this important point." Alternatively, please measure other pathways that regulate TFR1.

Thank you for this excellent suggestion. The sentences suggested by the reviewer have been added to the end of 4th paragraph of the Discussion section in the revised manuscript.

6) Aconitase is involved in iron responsiveness regulation in other tissues (Bullock Blood 2010). As a consequence, using aconitase as a read out of oxidative stress when iron is altered between the 2 groups is insufficient. Please remove the phrase "indicating increased oxidative stress" in the last sentence of the Results section.

According to the reviewer’s comment, we have removed the phrase “indicating increased oxidative stress” in the revised manuscript.

7) Please add "Further studies… increased TFR2 in the aged brain contributes to increased brain-derived hepcidin expression or alternatively contributes to iron accumulation with age." to the TFR2 part of the Discussion section.

Thank you for this suggestion. This is now corrected in the paper.

8) Question about muscle was intended to understand how data in muscle is relevant in a paper about the brain. This remains unclear.

As we mentioned in the introduction section, whether iron accumulation occurs in all organs or is specific to one organ was unclear despite aging has been shown to be associated with iron accumulation in various organs and tissues. Thus, we assessed aging-associated iron changes not only in the liver and the brain, but also in the skeletal muscle in the present study. We hope that the reviewer agrees that we show the iron data in the skeletal muscle in this paper.

9) Please add components of mouse strain and gender and haematologica reference for iron diet to the methods section.

We have added the information suggested by the reviewer in the method section of the revised manuscript.

Reviewer #2 (Recommendations for the authors):

The authors have addressed a number of concerns, including identifying the region of brain studied and validating the enriched mitochondrial fraction. Identifying the cell types with increased Hamp1/Hepc mRNA and protein and decreased Fpn protein is important, but this has not been done. In addition, the important finding of decreased Fpn is shown in a Western with an N of only 3. This is inadequate, especially with an outbred strain and documented high variability in Hamp1/Hepc levels in Figure 2. The N for the Fpn Western should be increased, and should be correlated with Hamp1/Hepc levels.

We appreciate the reviewer’s careful concern about the limitations of our experimental results. We acknowledge that the experiments proposed by the reviewer are important. However, due to the limited amount of sample and the time it takes to age these mice, it is unfortunately difficult to perform the additional experiments or analyses that are suggested by the reviewer. Instead, as recommended by the editor, the statement of "Since the results of this study are based on a small number of immunoblot samples using single-gender of mice, additional studies are needed to confirm these findings in a larger cohort of mice of both genders" has been added to the 2nd paragraph of the Discussion section in the revised manuscript as the limitation. We hope that the reviewer would agree with our response.

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    Figure 1—source data 1. Full-length images of immunoblots shown in Figure 1.
    elife-73456-fig1-data1.pptx (2.6MB, pptx)
    Figure 1—source data 2. Iron measurement – Original data of iron measurements shown in Figure 1.
    elife-73456-fig1-data2.xlsx (16.1KB, xlsx)
    Figure 1—source data 3. Densitometry – Original data of densitometric quantifications of immunoblots shown in Figure 1.
    elife-73456-fig1-data3.xlsx (10.8KB, xlsx)
    Figure 2—source data 1. Full-length images of immunoblots shown in Figure 2.
    elife-73456-fig2-data1.pptx (1.3MB, pptx)
    Figure 2—source data 2. qRT-PCR liver – Original data of qRT-PCR in the liver shown in Figure 2.
    elife-73456-fig2-data2.xlsx (84.6KB, xlsx)
    Figure 2—source data 3. qRT-PCR brain – Original data of qRT-PCR in the brain shown in Figure 2.
    elife-73456-fig2-data3.xlsx (85.8KB, xlsx)
    Figure 2—source data 4. Densitometry – Original data of densitometry quantifications of immunoblots shown in Figure 2.
    elife-73456-fig2-data4.xlsx (10.4KB, xlsx)
    Figure 3—source data 1. Full length images of immunoblots shown in Figure 3.
    elife-73456-fig3-data1.pptx (5.9MB, pptx)
    Figure 3—source data 2. Densitometry – Original data of densitometry quantifications of immunoblots shown in Figure 3.
    elife-73456-fig3-data2.xlsx (10.8KB, xlsx)
    Figure 3—source data 3. Aconitase assay – Original data of aconitase enzyme activities shown in Figure 3.
    elife-73456-fig3-data3.xlsx (10.2KB, xlsx)
    Transparent reporting form
    elife-73456-transrepform1.docx (245.5KB, docx)

    Data Availability Statement

    All data generated or analysed during this study were included as figures in the paper. All of the original datasets are included in the manuscript and supporting files.

    Articles from eLife are provided here courtesy of eLife Sciences Publications, Ltd

    RESOURCES