Mapping of mitochondrial ferritin in the brainstem of Macaca fascicularis

Neuroscience 328 (2016) 92–106

MAPPING OF MITOCHONDRIAL FERRITIN IN THE BRAINSTEM OF MACACA FASCICULARIS MINGCHUN YANG, a,b HONGKUAN YANG, a,b HONGPENG GUAN, a,b JEAN-PIERRE BELLIER, a SHIGUANG ZHAO b* AND IKUO TOOYAMA a* a

Molecular Neuroscience Research Center, Shiga University of Medical Science, Seta Tsukinowa-cho, Otsu 520-2192, Japan

b

Department of Neurosurgery, The First Affiliated Hospital of Harbin Medical University, Harbin Medical University, Harbin, Heilongjiang Province 150001, China

of FtMt which we found in the brainstem implies possible involvement of FtMt in several physiological mechanisms, especially in the catecholaminergic neurons, and the possibility of significant involvement in neurodegenerative disease. Ó 2016 The Author(s). Published by Elsevier Ltd on behalf of IBRO. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/ by-nc-nd/4.0/).

Key words: mitochondrial ferritin, iron, brainstem, catecholaminergic neuron, neurodegenerative disease.

Abstract—Mitochondrial ferritin (FtMt), a recently-studied iron storage protein, which we suspect is an important defense against oxidative stress in neurons and elsewhere. The 242-amino acid FtMt precursor protein is cleaved to mature protein (of molecular weight about 22-kDa) in the mitochondrial matrix. Compared with the ubiquitously expressed traditional ferritin (H-ferritin and L-ferritin), FtMt has been found in fewer locations including the testis, heart and brain. Previous studies have reported that the expression of FtMt in mouse and human brain is predominantly localized to neurons and partly to glial cells, and FtMt exerts protective effects on neurons by maintaining normal function and regulates apoptosis in Alzheimer’s disease and Parkinson’s disease. To find out the function of FtMt in neurodegenerative disease, we had a novel antibody made against human FtMt and characterized it via Western blot analysis, immunoabsorption testing, and double immunofluorescence histochemistry. Then we used this new FtMt antibody to map the distribution of FtMt in the monkey brainstem. We demonstrated widespread distribution of FtMt immunoreactivity throughout the monkey brainstem, with variable staining intensity. FtMt immunoreactivity was observed in the extrapyramidal system, sensory trigeminal nerve nuclei, some motor nuclei including ambiguous nucleus, dorsal motor nucleus of the vagus and hypoglossal nucleus, and some dorsal column nuclei such as the gracile nucleus and cuneate nucleus. In addition, double immunohistochemical stainings confirmed that FtMt immunoreactivity was co-localized with catecholaminergic neurons in the locus coeruleus (63.64%), substantia nigra pars compacta (69.18%), and ventral tegmental area (56.89%). The distribution

INTRODUCTION Mitochondrial ferritin (FtMt), a recently-identified iron storage protein, is preferentially located in mitochondria (Levi et al., 2001). FtMt is expressed as a precursor protein (242-amino-acid long, molecular weight around 30 kDa). It has 79% homology with H-chain ferritin (FTH) and ferroxidase but lacks functional iron-responsive elements (IREs). In the mitochondrial matrix, this precursor protein is cleaved to an approximately 22-kDa mature FtMt protein, functionally and structurally comparable to the cytosolic ferritins (Corsi et al., 2002; Drysdale et al., 2002; Levi and Arosio, 2004). FtMt assembles into 24-mer homopolymer shells which can act as ferroxidase and bind iron to the mitochondrial matrix. These characteristics make it highly likely that FtMt may protect against iron toxicity, at least within the mitochondria (Levi et al., 2001; Corsi et al., 2002; Levi and Arosio, 2004; Bou-Abdallah et al., 2005). This would be consistent with our previous report that FtMt protected neuroblastoma cells from oxidative stress (Wang et al., 2011). Iron, which is important in electron-transfer reactions, is an essential transition metal in many physiological functions. Various processes rely on reversible redox cycling between ferric and ferrous ions (Aisen et al., 2001; Dlouhy and Outten, 2013). Normally, plenty of iron is stored in the brain, as it is necessary for generating ATP in brain mitochondria. However, considerable evidence indicates that iron tends to accumulate in the brain at older ages (Connor et al., 1990; Zecca et al., 2004; Zucca et al., 2015), and excess iron is dangerous for cells, since it will lead to levels of toxic reactive oxygen species (ROS) (Connor et al., 1992; Welch et al., 2002). In fact, one hallmark of age-related diseases is excess free iron when storage in ferritin fails; ferritin is probably the oldest known and most widespread molecule

*Corresponding authors. Tel/fax: +86-451-53629254 (S. Zhao). Tel: +81-77-548-2330; fax: +81-077-548-2331 (I. Tooyama). E-mail addresses: [email protected] (S. Zhao), kinchan@belle. shiga-med.ac.jp (I. Tooyama). Abbreviations: AD, Alzheimer’s disease; COX4, cytochrome c oxidase subunit IV; EDTA, ethylenediaminetetraacetic acid; FRDA, Friedreich’s ataxia; FTH, H-chain ferritin; FtMt, mitochondrial ferritin; G, granule layer; HEPES, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; IREs, iron-responsive elements; M, molecular layer; P, Purkinje cells; PBST, phosphate-buffered saline containing 0.3% Triton X-100; PD, Parkinson’s disease; RLS, restless legs syndrome; ROS, reactive oxygen species; SDS, sodium dodecyl sulfate; TH, tyrosine hydroxylase.

http://dx.doi.org/10.1016/j.neuroscience.2016.04.035 0306-4522/Ó 2016 The Author(s). Published by Elsevier Ltd on behalf of IBRO. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 92

M. Yang et al. / Neuroscience 328 (2016) 92–106

moderating iron concentrations (Arosio et al., 2008; Altamura and Muckenthaler, 2009). In cells, overloaded iron is stored primarily in the cytoplasm, but is metabolically active mostly in the mitochondria, which are the main source of cellular energy and intracellular ROS. Thus, mitochondrial iron homeostasis should be tightly regulated to limit oxidative damage (Levi et al., 2001; Welch et al., 2002; Huang et al., 2011). Exactly how iron is controlled in mitochondria remains unclear, but FtMt’s ability to hold iron seems likely to be significant. FtMt tends to be found in metabolically highly activity tissue such as the testes, liver, spleen, and brain, and it is believed to be important in these tissues (Drysdale et al., 2002; Levi and Arosio, 2004). In the brain, accumulating evidence indicates involvement of FtMt in some neurodegenerative diseases, notably Alzheimer’s disease (AD) and Parkinson’s disease (PD), as well as in restless legs syndrome (RLS) (Nie et al., 2005; Snyder et al., 2009; Wang et al., 2011; Yang et al., 2013). FtMt also seems to act as a neuroprotective protein to maintain normal neuronal function and control apoptosis (Shi et al., 2010; Wu et al., 2013). Despite that, little has been published about FtMt distribution in the brain. Snyder et al. reported that FtMt-immunoreactivity was detected in neurons and oligodendrocytes in the mouse brain. We also reported that FtMt mRNA was expressed mainly by neurons in the human cerebral cortex (Snyder et al., 2010; Wang et al., 2011). However, neither of those studies gave detailed information on FtMt distribution in the brain. It is important to clarify the distribution patterns of FtMt in the brainstem, due to many reports suggesting that neurodegenerative diseases are closely associated with certain brainstem nucleus. In this study, we characterized a new polyclonal antibody made against human and monkey FtMt by Western blot analysis, immunoabsorption testing, and double immunofluorescence histochemistry. Our results proved that it specifically recognizes FtMt, and has no cross reaction with FTH. Using this antibody, we mapped the presence of FtMt in the cynomolgus monkey (Macaca fascicularis) brainstem. This study also revealed the localization of FtMt in catecholaminergic neurons.

EXPERIMENTAL PROCEDURES Animals The protocols of animals for this study were approved by the Institutional Animal Care and Use Committee of Shiga University of Medical Science. All the monkeys lived in normal housing condition (according to the care and feeding regulations of Animal Experiment Center of Shiga University of Medical Science). For Western blot analysis, the brainstem sample was obtained from two euthanized female cynomolgus monkeys (3 years 10 months, 2.67 kg; 12 years 2 months, 5.52 kg). For immunohistochemistry, brains were removed from four female cynomolgus monkeys (age: 5–11 years; weight: 3.38–4.68 kg) following use for other non-pathological study by other researchers. All efforts were made to avoid animal suffering and to minimize the number of animals used.

93

Production and characterization of an antibody against human FtMt We choose a synthetic peptide with an added N-terminal Cys residue (CTLGNENKQN: Accession number BC034419 on GenBank), which corresponds to amino acid residues 234–242 of human FtMt, as an immunogen. Then we requested that the Medical & Biological Laboratory Co. Ltd. (Ina, Nagano, Japan) produce antibodies against human FtMt. Antisera were raised in rabbits via immunization with this immunogen conjugated to keyhole limpet hemocyanin. Two rabbits were immunized with this antigen. After each booster immunization, the antisera were collected, and the best one was purified by affinity chromatography. Antibody specificity was tested by Western blot analysis, immunoabsorption, and fluorescence microscopy as described below.

Western blot analysis FtMt antibody specificity was checked by Western blot. Brainstem tissue was homogenized in a glass homogenizer with RIPA buffer [20 mM HEPES, 150 mM NaCl, 2 mM EDTA, 1% Nonidet P-40, 1% deoxycholic acid, and 0.1% sodium dodecyl sulfate (SDS) at pH 7.4] containing protease inhibitors [AEBSF: Aprotinin, Bestatin, E64, EDTA and Leupeptin (Sigma P2714), 1 mL/1 mg of USP pancreatin; Sigma, St. Louis, MO, USA], as described previously (Bellier and Kimura, 2007) with slight modifications. Mitochondria were isolated with Mitochondria Isolation Kit for Tissue (Thermo Fisher Scientific, Waltham, MA, USA) according to manufacturer’s instructions. Protein concentrations were assayed using a protein dye assay reagent (Bio-Rad Laboratories, Hercules, CA, USA) and an Infinite M200 reader (Tecan, Ma¨nnedorf, Switzerland). Crude protein (approximately 5 lg), prestained precision protein standards (Bio-Rad Laboratories), and Protein Ladders Standards (Bio-Rad, California, USA) or Protein Ladder One Triple-color (Nacalai Tesque, Kyoto, Japan) were electrophoresed on a 15% SDS–polyacrylamide gel (Wako Pure Chemical Industries, Osaka, Japan) under reducing conditions and then transferred to a polyvinylidene difluoride membrane (Immobilon-P, Merck Millipore, Billerica, MA, USA). Membranes were incubated for 1 h at room temperature with 20% Blocking reagent-N102 (NOF Corp., Tokyo, Japan) in 25 mM TBS containing 0.1% Tween-20 (TBST; pH 7.4) to block non-specific protein binding sites, followed by an 8 h incubation with the polyclonal FtMt antibody (0.128 lg/mL), cytochrome c oxidase subunit IV (COX4) (1:1000; Millipore) or FTH (1:5000; Abcam, Tokyo, Japan) at 4 °C. After three 10-min washes with TBST, the membranes were reacted with a peroxidase-labeled anti-rabbit IgG (1:50,000; Jackson ImmunoResearch Laboratories, West Grove, PA, USA) for 1 h at room temperature. Peroxidase labeling was detected by chemiluminescence using the Super Signal West Pico Chemiluminescent Substrate (Thermo Fisher Scientific, Waltham, MA, USA).

94

M. Yang et al. / Neuroscience 328 (2016) 92–106

As a control, Western blots were incubated without the primary antibody. We also performed an immunoabsorption test as another control. For this test, the purified antibody was pre-incubated with FtMt peptide (0.187 lg/mL) or human FTH full-length recombinant protein (Alpha Diagnostic International, San Antonio, USA) (0.357 lg/mL) at 4 °C for 24 h, and the membranes were incubated as described above.

Tissue preparation Tissue preparation was performed essentially according to previous reports (Abdelalim and Tooyama, 2011; Bisem et al., 2012). Brain stem samples were removed from four M. fascicularis, as mentioned above. Brains were collected at separate times and processed separately. All were fixed immediately with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) for 3 days at 4 °C. Next, the samples were immersed for 4 days in 15% sucrose in 0.1 M PB with 0.1% sodium azide; the sucrose solution was daily changed. The brains were then stored in 15% sucrose solution at 4 °C until sectioning. Samples were cryosectioned at 20 °C into 20-lm serial coronal sections that were floated in 0.1 M phosphate-buffered saline containing 0.3% Triton X-100 (PBST) at pH 7.4, and then maintained in PBST with sodium azide at 4 °C.

Immunohistochemistry Immunohistochemical staining for FtMt was performed as previously described (Abdelalim and Tooyama, 2011). In brief, sections were steeped in 0.1 M PBST (pH 7.4) for 3 days at 4 °C before staining. Sections were incubated for 40 min in PBST containing 0.1% sodium azide and 0.3% H2O2 at room temperature to inhibit endogenous peroxidase. After three 10-min washes in PBST, sections were incubated for 1 h with PBST containing 2% BSA. Next, the sections were incubated with purified FtMt antibody (0.128 lg/mL) for 3 days at 4 °C. After washing, the sections were incubated with biotinylated anti-rabbit IgG (1:3000; Vector Laboratories, Burlingame, CA, USA) for 1 h at room temperature. Finally, sections were incubated with ABC complex (1:3,000; ABC Elite; Vector Laboratories) for 1 h at room temperature, then the sections were developed in 0.02% 3,3-diamine-benzidine tetrahydrochloride with 0.3% nickel ammonium sulfate in 50 mM Tris–HCl (pH 7.6) containing 0.005% hydrogen peroxide. PBST was used to wash the sections between steps. For the immunoabsorption test, the FtMt antibody was pre-incubated with FtMt peptide (0.187 lg/mL) for 24 h at 4 °C, and the sections were stained as described above.

Mapping Mapping of FtMt distribution in the monkey midbrain, pons, and medulla oblongata was performed using camera lucida as described previously (Abdelalim et al., 2007; Bisem et al., 2012).

Immunofluorescence staining Tests were done in some midbrain sections to compare localization of FtMt and tyrosine hydroxylase (TH)-positive neurons, using double labeling for FtMt and TH, with fluorescently labeled secondary. To verify FtMt localization in the mitochondria, similar double labeling was done using FtMt and mitochondrial marker COX4 (a terminal enzyme of the mitochondrial electron transport chain) with fluorescently labeled secondary antibodies. Some sections were incubated with a mixture of FtMt polyclonal antibody (0.128 lg/mL) and a monoclonal antibody against TH (1:5000; Millipore) or COX4 (1:500; Millipore) at 4 °C overnight. After washing, sections were incubated in a mixture of Alexa Fluor 488-labeled anti-rabbit and Alexa Fluor 647-labeled anti-mouse secondary antibodies (1:500 dilutions; Molecular Probes Inc., Eugene, OR, USA) for 2 h at room temperature. After washing three times in phosphate-buffered saline (PBS), the sections were mounted on coated glass slides. Fluorescence was detected using a scanning laser confocal microscope (TE2000-E, Nikon, Tokyo, Japan). Single optical slice images were taken using 10, 20, or 40 Plan-Neofluor dry objective lenses. Micrographs Digital images taken with a Nikon-D90 digital camera (Tokyo, Japan) on an Olympus Microscope BX50 (Tokyo, Japan) were adjusted only for the brightness and contrast using Adobe Photoshop, with no further image manipulation. To add labels, an open source vector graphics editor named Inkscape was used.

RESULTS Characterization of the FtMt antibody The FtMt antibody was characterized by Western blot analysis of proteins, isolated mitochondria and cytosolic fractions extracted from the monkey midbrain, as well as an immunoabsorption test. In the Western blot analysis, the FtMt antibody detected a single band around 22 kDa, consistent with the known molecular weight of FtMt (arrow in Fig. 1A). Without the primary antibody, no such bands were observed; nor were they seen after the antibody was pre-absorbed with FtMt peptide (Fig. 1A). The antibody detected FtMt in mitochondria fraction (arrow in Fig. 1B), but not in cytosolic fraction (Fig. 1B). In addition, this antibody did not recognize FTH, as bands representing FtMt and FTH (Fig. 1C) were observed at different molecular weights. When the antibody was pre-absorbed with human FTH full-length recombinant protein, the signal did not reduce compared to the non-absorbed FtMt antibody (Fig. 1D). However, no bands were observed when FTH antibody was preincubated with this FTH recombinant protein (Fig. 1E). In tissues, this antibody stained neuronal cell bodies in specific brain regions (Fig. 1F). The staining was abolished when FtMt antibody pre-absorbed with FtMt peptide was used (Fig. 1G).

M. Yang et al. / Neuroscience 328 (2016) 92–106

95

Fig. 1. FtMt antibody characterization. (A) Western blot analysis of crude protein extract from a monkey brainstem, using our rabbit polyclonal antiFtMt antibody (FtMt+). Adjacent lanes were loaded with equivalent amounts of protein. The antibody recognized a single band at approximately 22 kDa (arrow). No bands were observed without primary antibody (FtMt ) or when antibody pre-absorbed with FtMt peptide was used (Pre-ab). (B) The antibody recognized FtMt in mitochondrial fraction (arrow in mito), but not in cytosolic fraction of monkey brainstem (cyto). (C) The antibody did not cross-react with monkey H-ferritin. FtMt and FTH bands were observed at different molecular weights. (D) The FtMt band signal was undiminished when antibody was pre-absorbed with human FTH full-length recombinant protein. (E) As a positive control, FTH signal was abolished following pre-incubation of the antibody with the FTH recombinant protein. (F) Immunostaining of FtMt in the oculomotor nucleus of the monkey midbrain, using our rabbit polyclonal anti-FtMt antibody. (G) No positive structures were detected when the antibody was pre-absorbed with FtMt peptide. Scale bars = 100 lm (C, D).

Localization of FtMt immunoreactivity in the monkey brainstem Western blotting of protein from mitochondria and cytosolic fractions revealed signal for COX4 only in total crude protein and mitochondria fraction extracted from monkey brainstem, but no band in the cytosolic fraction (Fig. 2A), indicating successful isolation of mitochondrial and cytoplasmic fractions from monkey brainstem. In the oculomotor nucleus (III), FtMt immunoreactivity was detected mainly in the cytoplasm of neurons. Confocal microscopic analysis revealed punctate FtMt positivity in the cytoplasm (Fig. 2B). Double immunofluorescent staining with the mitochondrial marker COX4 confirmed that FtMt was colocalized to COX4-positive mitochondria (Fig. 2D).

Distribution of FtMt in the monkey brainstem Widespread distribution of FtMt immunoreactivity was observed throughout the cynomolgus monkey brainstem in some neuronal structures, and some vessels were also weakly stained. FtMt immunoreactivity distribution seemed the same in all four monkey brainstems which we tested. Preferential FtMt staining was observed in neuronal cells, judging by simple morphological criteria (the size, shape and location of cells). The apparent staining of FtMt-positive neurons varied in intensity and density between regions, as summarized in Table 1. Fig. 3 represents the FtMt immunoreactivity distribution in the midbrain (Fig. 3A, B), pons (Fig. 3C–E), and medulla oblongata (Fig. 3E–G), drawn using a camera lucida. As shown in Fig. 3, the highest density of

96

M. Yang et al. / Neuroscience 328 (2016) 92–106

XII). The widespread distribution of FtMt suggests that FtMt is involved in various brain functions. Distribution of FtMt in the midbrain

Fig. 2. FtMt localized in mitochondria. (A) Western-blot analysis confirmed isolation of the mitochondrial fraction and cytosolic fraction from monkey brainstem. COX4 (arrows) was observed in total crude protein (total) and mitochondrial fraction (mito), but not in the cytosolic fraction (cyto). Confocal microscopic images of double immunofluorescent histochemistry for FtMt (B), Cox4 (B), and merged staining (C). Scale bars = 5 lm.

FtMt-immunoreactive structures was observed in the medulla oblongata, followed by the midbrain. These include catecholaminergic neurons (VTA, SNc, LC), sensory systems (MeV, Sp5, GR, LCU, and MCU), the extrapyramidal system (SC, IC, RN, PN, LR, LVe, MVe, DR, and MR) and some motor nuclei (AMB, DMNV, and

In the midbrain, FtMt immunoreactivity was seen in the cytoplasm and neuronal processes (arrowheads in Fig. 4) in the SNc, VTA, tectum, EW, III, RN, MGB, IV, and DR. FtMt immunoreactivity of varying intensity was observed in neurons in the SNc (Fig. 4A). FtMt immunoreactivity was primarily observed in pigmented cells (star), with positive staining in only a few nonpigmented cells (hollow star). Some FtMt immunoreactivity was observed in nuclei (hollow arrows) and fibers (hollow arrowheads). FtMt-immunoreactive neurons extended to the VTA (Fig. 4A). Weakly stained FtMt-immunoreactive neurons were seen in the superior and inferior colliculus (Fig. 4B). Some ependymal cells showed intense FtMt staining (Fig. 4B). FtMt-immunoreactive neurons with moderately intense staining were observed in the EW and III (Fig. 4C). Mild FtMt immunoreactivity was seen in the RN (Fig. 4C). Many closely-packed olivary FtMt-immunoreactive neurons were observed in the MGB (Fig. 4D). Some vessels were also weakly stained (arrows in Fig. 4D). In the caudal level, FtMt-immunoreactive neurons were found in the ventral side of the DR (Fig. 4E, F) and IV (Fig. 4E, G). Distribution of FtMt in the pons FtMt immunoreaction was observed in neurons of several regions (arrowheads in Fig. 5), including LVe, VI, MR, VR,

Table 1. Summary and abbreviations of brain regions where FtMt-positive stainings Region and Abbreviations Midbrain Central gray (CG) Substantia nigra pars compacta (SNc) Substantia nigra pars reticular (SNr) Red nucleus (RN) Ventral tegmental area (VTA) Oculomotor nucleus (III) Interpeduncular nucleus (IP) Central tegmental field (CTF) Edinger–Westphal nucleus (EW) Dorsal raphe nucleus (DR) Medial geniculate body (MGB) Superior colliculus (SC) Inferior colliculus (IC) Trochlear nucleus (IV) Pons Pedunculopontine tegmental nucleus (PPT) Pontine reticular nucleus (RTP) Pontine nuclei (PN) Locus coeruleus (LC) Mesencephalic trigeminal nucleus (Me5) Motor trigeminal nucleus (Mo5) Median raphe nucleus (MR) Ventral raphe nucleus (VR) Nucleus of trapezoid body (NTB)

Density

Intensity

1 3 – 1 2 3 – 3 ++ 2 3 3 3 2

+ ++ – ++ + ++ – + ++ ++ ++ + + ++

2 2 2 3 1 2 2 2 1

++ ++ ++ + +++ ++ ++ + +

Density of staining: 1, low; 2, intermediate; 3, high. Intensity of staining: +, weak; ++, moderate; +++, strong.

Region and Abbreviations

Density

Intensity

Abducens nucleus (VI) Lateral vestibular nucleus (LVe) Parabrachial nucleus (PB) Medulla oblongata Medial vestibular nucleus (MVe) Gracilis nucleus (GR) Lateral cuneate nucleus (LCU) Medial cuneate nucleus (MCU) Raphe obscurus nucleus (RO) Medial longitudinal fasciculus (MLF) Spinal trigeminal nucleus (Sp5) Ambiguus nucleus (AMB) Supraspinal nucleus (SSp) Lateral reticular nucleus (LR) Inferior olivary nucleus (IO) Nucleus of the solitary tract (SOL) Dorsal motor nucleus of vagus (DMNV) Hypoglossal nucleus (XII) Cerebellum Deep cerebellar nuclei (CBN) Molecular layer (M layer) Purkinje cell layer (P cells) Granular cell layer (G cells)

2 3 —

++ ++ —

3 2 3 3 3 3 3 3 2 2 3 3 3 3

++ + +++ +++ + ++ ++ +++ +++ ++ ++ + +++ +++

3 3 1 3

++ ++ +++ ++

M. Yang et al. / Neuroscience 328 (2016) 92–106

97

Fig. 3. Distribution of FtMt immunoreactivity in the brainstems of cynomolgus monkeys, drawn from representative sections using a camera lucida. Directional reference orientation is shown at the bottom, indicating the orientation: S: Superior, I: Inferior, R: Right, L: Left, F: Frontal, and C:Caudal.

98

M. Yang et al. / Neuroscience 328 (2016) 92–106

Fig. 4. FtMt immunoreactivity in the midbrains of cynomolgus monkeys. (A–G) FtMt-immunoreactivity in specific regions of the midbrain. Arrowheads indicate immunoreactivity in the cytoplasm and neuronal processes. Higher magnification images of FtMt immunoreactivity are shown in boxed areas (A–E). (A) The substantia nigra (SN) and ventral tegmental area (VTA). A Star indicate pigmented cells, a hollow star indicate non-pigmented cells, hollow arrows indicate nuclei, and hollow arrowheads indicate fibers. (B) Superior colliculus (SC) and ependymal cells. (C) Edinger–Westphal (EW), oculomotor (III), and red nuclei (RN). (D) Medial geniculate body (MGB) and blood vessels (BV) (arrows). (E) Dorsal raphe (DR) and trochlear (IV) nucleus. (F, G) High-magnification images of FtMt immunoreactivity in the IV and DR. Scale bars = 500 lm (A, B, C, E), 200 lm (D), 50 lm (F, G).

RTP, PPT, PN, Mo5, Me5, and LC. Spindle-shaped and multipolar immunoreactive neurons were seen in the LVe (Fig. 5A). Strongly stained FtMt-immunoreactive neurons were observed in the VI under the fourth ventricle (Fig. 5A). Neurons in the MR and RTP exhibited intense FtMt staining (Fig. 5B). In the VR, some weekly immunoreactive perikarya were seen (Fig. 5C). FtMt immunoreactivity was also observed in

the neuronal cytoplasm of the PPT (Fig. 5D). Moderately intense staining of fusiform immunoreactive neurons was seen in the PN (Fig. 5D). Mildly stained FtMt-immunoreactive perikarya were observed in the Mo5 (Fig. 5E). Several FtMt-positive neurons were detected in the Me5, with intense and punctate staining (Fig. 5F). In addition, typical mitochondrial staining was detected in the LC (Fig. 5F).

M. Yang et al. / Neuroscience 328 (2016) 92–106

99

Fig. 5. FtMt immunoreactivity in the pons of cynomolgus monkeys. (A–F) FtMt-immunoreactive neurons in the pons. Higher magnification images of FtMt immunoreactivity are shown in the boxed areas. Arrowheads indicate immunoreactivity in the cytoplasm and neuronal processes. (A) Lateral vestibular nucleus (LVe) and abducens nucleus (VI). (B) Median raphe nucleus (MR) and nucleus reticularis tegmenti pontis (RTP). (C) Ventral raphe nucleus (VR). (D) Pontine nuclei (PN) and pedunculopontine tegmental nucleus (PPT). (E) Motor trigeminal nucleus (Mo5). (F) Locus coeruleus (LC) and mesencephalic nucleus of the trigeminal nerve (Me5). Scale bars = 500 lm (A–D), 100 lm (E, F).

Distribution of FtMt in the medulla oblongata Some regions in the medulla oblongata exhibited FtMt immunoreactivity, including MCU, GR, LCU, MLF, RO, Sp5, AMB, SSp, LR, MVe, SOL, DMNV, XII, and IO (arrowheads in Figs. 6 and 7A–D). Highly immunoreactive and variably shaped neurons were observed in the MCU (Fig. 6A). In the GR, mild punctate FtMt immunoreaction was detected in neurons (Fig. 6B). Strong FtMt-immunoreactive neurons were seen in the LCU (Fig. 6C). Moderately FtMt-immunoreactive neurons and neuronal short processes were visible in the MLF (Fig. 6D). Mildly oval-shaped stained neurons were found in the RO (Fig. 6D). Moderate intensity, punctate FtMt-stained neurons were observed in the Sp5 (Fig. 6E). In the AMB, clusters of strongly FtMt-immunoreactive neurons were detected (Fig. 6E). In the SSp, intense fusiform FtMt immunoreactivity appeared in elongated neuron

cell body and their processes (Fig. 6E). In the LR, FtMt-immunoreactive perikarya were also observed (Fig. 6E). A large number of uniformly distributed FtMt staining was seen throughout the IO (Figs. 6E and 7D). In the MVe, variably shaped FtMt-immunoreactive neurons were observed (Fig. 7A). Abundant areas of weak, punctate FtMt immunoreactivity was observed in the SOL (Fig. 7B). In the DMNV and XII, obvious punctate perinuclear staining was detected in FtMtimmunoreactive neurons (Fig. 7C).

Distribution of FtMt-immunoreactive structures in the cerebellum We also observed intense FtMt immunoreactivity in the cerebellum (Fig. 7E). A large number of closely packed areas of staining were detected in the granule layer (G) of the cerebellar gray matter, and moderately intense

100

M. Yang et al. / Neuroscience 328 (2016) 92–106

Fig. 6. FtMt immunoreactivity in the medulla oblongata of cynomolgus monkeys. (A–E) Immunoreactive neurons in the specific region of medulla oblongata. Boxed areas show higher magnification images of FtMt immunoreactivity. Arrowheads indicate punctate immunoreactivity in the cytoplasm and neuronal processes. (A) Medial cuneate nucleus (MCU). (B) Gracile nucleus (GR). (C) Lateral cuneate nucleus (LCU). (D) Medial longitudinal bundle (MLB) and raphe obscurus nucleus (RO). (E) Ambiguous nucleus (AMB), inferior olivary nucleus (IO), lateral reticular nucleus (LR), spinal trigeminal nucleus (Sp5), and supraspinal nucleus (SSp). Scale bars = 500 lm (A, D, E), 200 lm (B, C).

M. Yang et al. / Neuroscience 328 (2016) 92–106

101

Fig. 7. FtMt immunoreactivity in the medulla oblongata and cerebellum of cynomolgus monkeys. (A–D) FtMt-immunoreactive neurons in the medulla oblongata. Boxed areas are higher magnification images of FtMt immunoreactivity. Arrowheads indicate punctate immunoreactivity in the cytoplasm and neuronal processes. (A) Medial vestibular nucleus (MVe). (B) Solitary nucleus (SOL). (C) Dorsal motor nucleus of the vagus (DMNV) and hypoglossal nucleus (XII). (D) Inferior olivary nucleus (IO), (E–G) FtMt staining in the cerebellum. (F, G) Higher magnification image showing a large number of closely packed areas of FtMt immunoreactivity in granule cells (G cells), along with immunoreactive cells in the molecular layer (M layer). Strong staining was present in Purkinje cells (P, arrows). FtMt staining was also observed in deep cerebellar nuclei (arrowheads). Scale bars = 500 lm (A, D, E), 200 lm (B, C), 100 lm (F, G).

staining was visible in the molecular layer (M). Distinct FtMt was sparse but consistently observed in Purkinje cells (P) (Fig. 7E, F). In addition, FtMt-immunoreactive

neurons were equally distributed in the medial, interposed, and lateral nucleus of the cerebellum (Fig. 7E, F).

102

M. Yang et al. / Neuroscience 328 (2016) 92–106

Localization of FtMt in catecholaminergic neurons of the monkey brainstem To test the possibility that catecholaminergic neurons express FtMt, double immunofluorescence histochemistry was employed. We subjected monkey brainstem samples to double immunostaining for FtMt and TH. Our results revealed relatively strong co-localization of FtMt- and TH-positive neurons (white arrows) in the VTA (Fig. 8A–C), SNc (Fig. 8D–F), and LC (Fig. 8G–I). Co-localization occurred mainly in the perikarya. Particularly strong FtMt staining was observed in the Me5 (Fig. 8I), consistent with the previous immunohistochemical result (Fig. 5F). The percentage of co-FtMt and TH immunoreactivity co-localization is shown in Table 2. A high percentage of neurons were doubly stained relative to the number of TH-positive neurons in the VTA (56.89%), SN (69.18%), and LC (63.64%), and the percentages of double-positive neurons to total FtMt-positive neurons were 71.73% in VTA, 66.40% in SNc, and 75.66% in LC.

Table 2. Percentage of neurons double-stained for tyrosine hydroxylase (TH) and mitochondrial ferritin (FtMt) relative to the total populations of TH-positive neurons (upper row) and FtMt-positive neurons (lower row) in the brainstems of cynomolgus monkeys Region

VTA (%)

SNc (%)

LC (%)

Percentage of double-positive neurons among total TH-positive neurons Percentage of double-positive neurons among total FtMt-positive neurons

56.89

69.18

63.64

71.73

66.40

75.66

DISCUSSION Characterization of the FtMt antibody In the present study, when we using Western blotting test the expression of FtMt in the monkey brainstem, the FtMt antibody detected a single band only in crude protein and mitochondrial fraction extracted from monkey brainstem with an approximate molecular weight of 22 kDa. This band showed no decrease when the FtMt antibody was

Fig. 8. Double immunofluorescent staining for FtMt (green) and TH (red), showing co-localization of FtMt-positive and TH-positive neurons in the ventral tegmental area (VTA) (A–C), substantia nigra (SN) (D–F), and locus coeruleus (LC) (G–I). (C, F) Numerous neurons in the VTA and SN exhibited co-localization of FtMt and TH immunoreactivity. (I) Nearly all FtMt-positive neurons in the LC were also TH immunoreactive. High-intensity FtMt staining was observed in the Me5 (white arrowheads). White arrows indicate double-positive neurons (C, F, I). Scale bars = 50 lm (A–F), 25 lm(G–I).

M. Yang et al. / Neuroscience 328 (2016) 92–106

pre-absorbed with human FTH recombinant protein, but it was not observed when the FtMt antibody was omitted or pre-absorbed with antigenic peptide (CTLGNENKQN). Meanwhile, this band was observed in different molecular weight with FTH, thus the antibody does not recognize monkey H-ferritin. Moreover, in another study conducted by our group, this FtMt antibody was not found to cross-react with human FTH in HEK293 cells (Yang et al., 2015). In addition, the immunohistochemical study using the antibody showed that FtMt-positive neurons were observed in specific brain regions, and they were abolished by the antibody which was pre-absorbed with FtMt peptide. Furthermore, double immunofluorescent histochemistry for FtMt and COX4 confirmed the mitochondrial localization of FtMt. These results indicate that this antibody recognized FtMt specifically in the monkey brain. Distribution of FtMt in the monkey brainstem This is the first study to map FtMt protein distribution in the brainstem. We demonstrated that FtMt immunoreactivity was observed preferentially in neuronal cells of the brainstem, where it was observed in the mitochondria and cytoplasm of somata and neuronal processes. The neuronal localization of FtMt is in good agreement with previous reports (Snyder et al., 2010; Wang et al., 2011), even if some oligodendrocytes were reported to be also positive for FtMt. In this study, except TH-double-labeling, we did not perform further doublelabeling analysis to distinguish other cell types. We suspect that FtMt expression in glial cells is either absent or very low in normal physiological conditions, but that cannot rule out the possibility of an increase under pathological conditions. Our previous study showed that FtMt mRNA was mainly localized in neurons of normal human brain, but some glial cells expressing FtMt mRNA were detected in the brains of AD patients (Wang et al., 2011). Since FtMt is an iron storage protein, one may ask whether its distribution is affected by the iron status. Indeed, there is accumulating evidence demonstrating that iron is homogenously distributed in the normal adult human brain, with higher levels in the midbrain (i.e., SN) and cerebellum than in the pons and medulla (House et al., 2012; Ramos et al., 2014; Zucca et al., 2015). For example, FtMt immunoreactivity in the iron-rich SN did not appear to be more robust than in other nuclei of the midbrain, such as the III. These findings suggest that distribution of FtMt is not affected by the iron content in the brainstem. This is in accord with a previous report in which FtMt expression was indicated to be ironindependent because of the lacking IREs (Levi et al., 2001). A previous study reported that FtMt expression may be influenced by the mitochondrial density, but not by iron content (Snyder et al., 2009). Our results provide evidence for the hypothesis, as the differences in the intensity and form of FtMt immunostaining appear to be consistent with mitochondrial density and neuronal activity. Interestingly, a terminal enzyme of the mitochondrial electron transport chain, COX4, is considered an indicator of the metabolic abilities and functional activity of neurons (Wong-Riley, 1989). Some studies observed

103

positive staining for Cox4 and ubiquitous mitochondrial creatine kinase, an important enzyme for neuronal energy metabolism, in the human brainstem. The stained neurons appeared in the DR, IC, III, IV, Me5, PAG, SC, SNc, and RN of the midbrain; LC, LVe, MR, Mo5, PN, Sp5, VI, and VII of the pons; and AMB, CU, DMNV, GR, IO, LR, MLF, MVe, SOL, and XII of the medulla (Liu and Wong-Riley, 2003, 2005; Lowe et al., 2013). Overall, our results regarding the distribution of FtMt-positive neurons in the brainstem were very similar to the mitochondrial distributions described in earlier reports, particularly in terms of the intensity and pattern of staining. In addition, FtMt-staining in the M, G, and P of the cerebellum was also consistent with mitochondrial distribution (Nie and Wong-Riley, 1996). The main role of the mitochondria is to supply cellular energy; accordingly, mitochondrial density represents the energy consumption and demand of a cell, as well as its activity level (McBride et al., 2006). Mitochondrial density is among the main determinants of the metabolic rate, as this factor indicates the maximum rate of energy conversion from stored material (Pathi et al., 2013). To satisfy their bioenergetic needs, neurons require high level of mitochondrial oxidative phosphorylation, which is a major cellular source of ROS. This may imply the ROS content is proportional to the mitochondrial density in neurons (Chance et al., 1979; Nicholls, 2008; Sanchez-Padilla et al., 2014). However, some pathological conditions can induce more ROS production than cellular antioxidant capacity, such as in the LC and SNc of PD patients (Dickson, 2012; Hauser and Hastings, 2013). Our previous in vitro study indicated that the increased mRNA and protein levels expression of FtMt may be induced by ROS and Amyloid beta (Wang et al., 2011). These suggest FtMt is not constitutively expressed, but its expression might change in pathologic conditions. Indeed, increased FtMt expression was observed in the temporal cortexes of patients, compared with normal individuals, and was also observed in the SN of PD or RLS patients (Nie et al., 2005; Snyder et al., 2009; Wang et al., 2011). Taken together, these findings suggest that FtMt expression is influenced by either mitochondrial density and functional activity of neurons. As the present study is restricted to 5–11-year-old normal disease-free M. fascicularis, we did not see direct evidence about this alteration or the expression of FtMt in different age monkeys. Further studies are needed to establish the age-related and pathological conditions induced expression of FtMt. FtMt in neurodegenerative diseases Many studies have implicated FtMt in some neurodegenerative diseases such as AD, PD, and RLS. Parkinsonism is a very common neurodegenerative disorder characterized by prominent bradykinesia and extrapyramidal signs. A key feature of PD is the build-up of a-synuclein in Lewy bodies (Dickson, 2012). Free iron is toxic, causing oxidation and cell damage, and uncontrolled iron levels tend to rise with aging (Connor et al., 1990; Zecca et al., 2004). Also, excess iron tends to be associated with neurodegenerative diseases, particularly AD and PD (Zecca et al., 2004; Kell, 2010). Recent

104

M. Yang et al. / Neuroscience 328 (2016) 92–106

reports described the detection of FtMt throughout the layers of the cerebral cortex and in the cerebellum, hippocampus, striatum, and SN (Shi et al., 2010; Snyder et al., 2010). Although FtMt has been thought to prevent cell death induced by ROS and neurotoxic proteins in patients with neurodegenerative disease by maintaining iron homeostasis (Nie et al., 2005; Snyder et al., 2009; Wang et al., 2011), there is little information about FtMt expression in brainstem regions in association with these diseases. In the present study, FtMt-immunoreactive neurons were recognized in the extrapyramidal system (SC, IC, RN, PN, LR, LVe, MVe, DR, and MR), some motor nuclei (AMB, DMNV, and XII) and some regions contain catecholaminergic neurons (SN, LC and VTA). The staining of FtMt appeared similar in these regions, with typical punctate mitochondrial staining. Evidence suggests that iron deposition increases with normal aging, is most strongly present in structures of the extrapyramidal system, and might be a cause to functional decline (Bartzokis et al., 2007). A new study shows that a-synuclein overexpression contributes to iron accumulation in neurons (Ortega et al., 2016). Studies have shown that Lewy bodies, which are composed of the protein a-synuclein, are most prevalent in the SN, VTA, PPT, AMB, DMNV, XII, SOL, LC, raphe nuclei, periaqueductal gray matter, and reticular formation in PD patients (Kingsbury et al., 2010; Kokubo et al., 2012; Seidel et al., 2015). In another study by our group, FtMt modulates a-synuclein expression via iron regulation in physiological conditions (submitted paper). In the present study, FtMt immunoreactivity was observed in the sensory trigeminal nerve nuclei (MeV, Sp5, and V) and dorsal column nuclei (GR, LCU, and MCU). Friedreich’s ataxia (FRDA) is another neuropathy associated with mitochondrial iron control: when the iron-controlling mitochondrial protein called frataxin is lacking, the resulting uncontrolled iron causes neural damage. Among the effects is a severe loss of secondary sensory neurons, including the spinal and principal trigeminal nuclei and, in particular, the mesencephalic trigeminal nucleus in the brainstem (Jitpimolmard et al., 1993), and neuronal atrophy in the GR and CU has been observed in FRDA (Koeppen and Mazurkiewicz, 2013). Being another mitochondrial iron-holding protein, FtMt may act to limit the damage in FRDA by modulating mitochondrial iron availability, and hence reducing oxidative damage (Zanella et al., 2008; Campanella et al., 2009). In general, the distribution of FtMt immunoreactivity seen in the present study provides support for previous studies and suggests that FtMt may be important in neurodegenerative diseases, especially Parkinsonism. Exactly what FtMt does in the brainstem (and elsewhere) in normal and abnormal physiology has yet to be clarified.

immunofluorescent staining for FtMt and TH confirmed that the percentages of double-positive neurons to total TH-positive neurons were 56.89% in VTA, 69.18% in SNc and 63.64% in LC. Interestingly, several studies have reported possible involvements of FtMt in midbrain dopaminergic neurons affected by neurological disorders such as PD (Shi et al., 2010) and RLS (Snyder et al., 2009). PD is well known as the loss of melanin-containing dopaminergic neurons in the SNc and to a lesser extent in the noradrenergic neurons in the LC. The neurons in these two regions may be preferentially destroyed by oxidative damage (Halliwell, 2006; Dickson, 2012). A study reported that cumulating oxidative stress may lead to a cellular energy shortage resulting from mitochondrial dysfunction (Nicholls, 2008). FtMt, by controlling iron in mitochondria, protects against oxidative stress to neurons through the Fenton reaction (Yang et al., 2013). Dopaminergic neurons of the midbrain include two important sets of projecting cells: the ‘‘A9” neurons projecting from the SN and the ‘‘A10” cells from the VTA (Grison et al., 2014). In PD, selective degeneration occurs in A9 neurons but not A10 cells. Normally, dopaminergic neurons in the SNc are significantly more sensitive to oxidative stress than are those in the VTA (Horowitz et al., 2011). However, abnormal A10 cell function has been linked to addiction, attention deficit, and schizophrenia (Grison et al., 2014). Interestingly, our previous study indicated that oxidative stress could increase the expression of both FtMt mRNA and protein (Wang et al., 2011). A previous study reported that some subpopulations of LC neurons are very sensitive to oxidative stress (de Oliveira et al., 2012). Although there is no direct evidence, it may be conjectured that more FtMt might be expressed in such LC neurons. Together our observations may imply the different expression of FtMt in the TH-positive neurons in the VTA, LC and SN related to regional differences in sensitivity to oxidative stress. These findings suggest FtMt may be important in catecholaminergic neurons.

CONCLUSION In the present study, we characterized a new human FtMt antibody, and with it created the first map of FtMt immunoreactivity in a monkey brainstem. This showed the widespread distribution of FtMt in various brainstem regions and co-localization with catecholaminergic neurons in the LC, SN, and VTA, suggesting that FtMt plays a significant role in catecholamine neurons and more importantly, it could be involved in neurodegenerative diseases such as Parkinsonism and ataxia. The map of FtMt immunoreactivity should give fundamental data to understand roles of FtMt in physiological and pathological brain functions.

AUTHOR CONTRIBUTIONS FtMt in the catecholaminergic neurons In the current study, we observed FtMt immunoreactivity in neurons of the VTA, SN, and LC which are the regions contains catecholaminergic neurons. Double

Designed the experiments: MC.Y., SG.Z., I.T. Performed the experiments: MC.Y. Contributed with material collection: MC.Y., HP. Antibody production: MC.Y., HK.Y., I.T.

M. Yang et al. / Neuroscience 328 (2016) 92–106

Analyzed the data: MC.Y., JP.B., SG.Z., I.T. Edited the figures: MC.Y. Wrote the paper: MC.Y., JP.B, I.T. Formatted/submitted the paper: I.T. Institutional support: I.T. All authors have approved the final manuscript.

STATEMENT OF INTERESTS All authors have no conflicts of interest to disclose. Acknowledgments—This study was supported by Grant-in-Aids for Scientific Research (B) (Grant Number 26290022, I.T.) from Japan Society for the Promotion of Science (JSPS), and a Grant-in-Aid for Scientific Research on Innovative Areas (‘‘Brain Environment”) (MEXT KAKENHI Grant Number 26640042, I.T.) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (MEXT). Dr. Mingchun Yang was supported by the Otsuka Toshimi Scholarship foundation. We thank all members of the laboratory who helped with this work and provided suggestions for brain mapping. The authors would like to thank Mr. T. Yamamoto of the Central Research Laboratory, Shiga University of Medical Science, for his excellent technical assistance with confocal microscopy.

REFERENCES Abdelalim EM, Tooyama I (2011) Mapping of NPR-B immunoreactibity in the brainstem of Macaca fascicularis. Brain Struct Funct 216:387–402. Abdelalim EM, Osman AH, Takada T, Torii R, Tooyama I (2007) Immunohistochemical mapping of natriuretic peptide receptor-A in the brainstem of Macaca fascicularis. Neuroscience 145:1087–1096. Aisen P, Enns C, Wessling-Resnick M (2001) Chemistry and biology of eukaryotic iron metabolism. Int J Biochem Cell Biol 33:940–959. Altamura S, Muckenthaler MU (2009) Iron toxicity in diseases of aging: Alzheimer’s disease, Parkinson’s disease and atherosclerosis. J Alzheimers Dis 16:879–895. Arosio P, Ingrassia R, Cavadini P (2008) Ferritins: a family of molecules for iron storage, antioxidation and more. Biochim Biophys Acta 1790:589–599. Bartzokis G, Tishler TA, Lu PH, Villablanca P, Altshuler LL, Carter M (2007) Brain ferritin iron may influence age- and gender-related risks of neurodegeneration. Neurobiol Aging 28:414–423. Bellier JP, Kimura H (2007) Acetylcholine synthesis by choline acetyltransferase of a peripheral type as demonstrated in adult rat dorsal root ganglion. J Neurochem 101:1607–1618. Bisem NJ, Takeuchi S, Imamura T, Abdelalim EM, Tooyama I (2012) Mapping of FGF1 in the medulla oblongata of Macaca fascicularis. Acta Histochem Cytochem 45:325–334. Bou-Abdallah F, Santambrogio P, Levi S, Arosio P, Chastten ND (2005) Unique iron binding and oxidation properties of human mitochondrial ferritin: a comparative analysis with human H-chain ferritin. J Mol Biol 347:543–554. Campanella A, Rovelli E, Santambrogio P, Cozzi A, Taroni F, Levi S (2009) Mitochondrial ferritin limits oxidative damage regulating mitochondrial iron availability: hypothesis for a protective role in Friedreich ataxia. Hum Mol Genet 18:1–11. Chance B, Sies H, Boveris A (1979) Hydroperoxide metabolism in mammalian organs. Physiol Rev 59:527–605. Connor JR, Menzies SL, St Martin SM, Mufson EJ (1990) Cellular distribution of transferrin, ferritin, and iron in normal and aged human brains. J Neurosci Res 27:595–611.

105

Connor JR, Menzies SL, St Martin SM, Mufson EJ (1992) A histochemical study of iron, transferrin, and ferritin in Alzheimer’s diseased brains. J Neurosci Res 31:75–83. Corsi B, Cozzi A, Arosio P, Drysdale J, Santambrogio P, Campanella A, Biasiotto G, Albertini A, Levi S (2002) Human mitochondrial ferritin expressed in HeLa cells incorporates iron and affects cellular iron metabolism. J Biol Chem 277: 22430–22437. de Oliveira RB, Gravina FS, Lim R, Brichta AM, Callister RJ, van Helden DF (2012) Heterogeneous responses to antioxidants in noradrenergic neurons of the Locus coeruleus indicate differing susceptibility to free radical content. Oxid Med Cell Longev 2012:820285. Dickson DW (2012) Parkinson’s disease and parkinsonism: neuropathology. Cold Spring Harb Perspect Med 2:a009258. Dlouhy AC, Outten CE (2013) The iron metallome in eukaryotic organisms. Met Ions Life Sci 12:241–278. Drysdale J, Arosio P, Invernizzi R, Cazzola M, Volz A, Corsi B, Biasiotto G, Levi S (2002) Mitochondrial ferritin: a new player in iron metabolism. Blood Cells Mol Dis 29:376–383. Grison A, Zucchelli S, Urzı` A, Zamparo I, Lazarevic D, Pascarella G, Roncaglia P, Giorgetti A, Garcia-Esparcia P, Vlachouli C, Simone R, Persichetti F, Forrest AR, Hayashizaki Y, Carloni P, Ferrer I, Lodovichi C, Plessy C, FANTOM Consortium, Carninci P, Gustincich S (2014) Mesencephalic dopaminergic neurons express a repertoire of olfactory receptors and respond to odorant-like molecules. BMC Genomics 15:729. Halliwell B (2006) Oxidative stress and neurodegeneration: where are we now? J Neurochem 97:1634–1658. Hauser DN, Hastings TG (2013) Mitochondrial dysfunction and oxidative stress in Parkinson’s disease and monogenic parkinsonism. Neurobiol Dis 51:35–42. Horowitz MP, Milanese C, Di Maio R, Hu X, Montero LM, Sanders LH, Tapias V, Sepe S, van Cappellen WA, Burton EA, Greenamyre JT, Mastroberardino PG (2011) Single-cell redox imaging demonstrates a distinctive response of dopaminergic neurons to oxidative insults. Antioxid Redox Signal 15:855–871. House E, Esiri M, Forster G, Ince PG, Exley C (2012) Aluminium, iron and copper in human brain tissues donated to the Medical Research Council’s Cognitive Function and Ageing Study. Metallomics 4:56–65. Huang J, Jones D, Luo B, Sanderson M, Soto J, Abel ED (2011) Iron overload and diabetes risk: a shift from glucose to Fatty Acid oxidation and increased hepatic glucose production in a mouse model of hereditary hemochromatosis. Diabetes 60:80–87. Jitpimolmard S, Small J, King RH, Geddes J, Misra P, McLaughlin J, Muddle JR, Cole M, Harding AE, Thomas PK (1993) The sensory neuropathy of Friedreich’s ataxia: an autopsy study of a case with prolonged survival. Acta Neuropathol 86:29–35. Kell DB (2010) Towards a unifying, systems biology understanding of large-scale cellular death and destruction caused by poorly liganded iron: Parkinson’s, Huntington’s, Alzheimer’s, prions, bactericides, chemical toxicology and others as examples. Arch Toxicol 84:825–889. Kingsbury AE, Bandopadhyay R, Silveira-Moriyama L, Ayling H, Kallis C, Sterlacci W, Maeir H, Poewe W, Lees AJ (2010) Brain stem pathology in Parkinson’s disease: an evaluation of the Braak staging model. Mov Disord 25:2508–2515. Koeppen AH, Mazurkiewicz JE (2013) Friedreich ataxia: neuropathology revised. J Neuropathol Exp Neurol 72:78–90. Kokubo Y, Taniguchi A, Hasegawa M, Hayakawa Y, Morimoto S, Yoneda M, Hirokawa Y, Shiraishi T, Saito Y, Murayama S, Kuzuhara S (2012) a-Synuclein pathology in the amyotrophic lateral sclerosis/parkinsonism dementia complex in the Kii Peninsula, Japan. J Neuropathol Exp Neurol 71:625–630. Levi S, Arosio P (2004) Mitochondrial ferritin. Int J Biochem Cell Biol 36:1887–1889. Levi S, Corsi B, Bosisio M, Invernizzi R, Volz A, Sanford D, Arosio P, Drysdale J (2001) A human mitochondrial ferritin encoded by an intron less gene. J Biol Chem 276:24437–24440.

106

M. Yang et al. / Neuroscience 328 (2016) 92–106

Liu Q, Wong-Riley MT (2003) Postnatal changes in cytochrome oxidase expressions in brain stem nuclei of rats: implications for sensitive periods. J Appl Physiol 95:2285–2291. Liu Q, Wong-Riley MT (2005) Postnatal developmental expressions of neurotransmitters and receptors in various brain stem nuclei of rats. J Appl Physiol 98:1442–1457. Lowe MT, Kim EH, Faull RL, Christie DL, Waldvogel HJ (2013) Dissociated expression of mitochondrial and cytosolic creatine kinases in the human brain: a new perspective on the role of creatine in brain energy metabolism. J Cereb Blood Flow Metab 33:1295–1306. McBride HM, Neuspiel M, Wasiak S (2006) Mitochondria: more than just a powerhouse. Curr Biol 16:R551–R560. Nicholls DG (2008) Oxidative stress and energy crises in neuron dysfunction. Ann N Y Acad Sci 1147:53–60. Nie F, Wong-Riley MT (1996) Mitochondrial- and nuclear-encoded subunits of cytochrome oxidase in neurons: differences in compartmental distribution, correlation with enzyme activity, and regulation by neuronal activity. J Comp Neurol 373:139–155. Nie G, Sheftel AD, Kim SF, Ponka P (2005) Overexpression of mitochondrial ferritin causes cytosolic iron depletion and changes cellular iron homeostasis. Blood 105:2161–2167. Ortega R, Carmona A, Roudeau S, Perrin L, Ducˇic´ T, Carboni E, Bohic S, Cloetens P, Lingor P (2016) a-Synuclein over-expression induces increased iron accumulation and redistribution in iron-exposed neurons. Mol Neurobiol 53:1925–1954. Pathi B, Kinsey ST, Locke BR (2013) Oxygen control of intracellular distribution of mitochondria in muscle fibers. Biotechnol Bioeng 110:2513–2524. Ramos P, Santos A, Pinto NR, Mendes R, Magalha˜es T, Almeida A (2014) Iron levels in the human brain: a post-mortem study of anatomical region differences and age-related changes. J Trace Elem Med Biol 28:13–17. Sanchez-Padilla J, Guzman JN, Ilijic E, Kondapalli J, Galtieri DJ, Yang B, Schieber S, Oertel W, Wokosin D, Schumacker PT, Surmeier DJ (2014) Mitochondrial oxidant stress in locus coeruleus is regulated by activity and nitric oxide synthase. Nat Neurosci 17:832–840. Seidel K, Mahlke J, Siswanto S, Kru¨ger R, Heinsen H, Auburger G, Bouzrou M, Grinberg LT, Wicht H, Korf HW, den Dunnen W, Ru¨b U (2015) The brainstem pathologies of Parkinson’s disease and dementia with Lewy bodies. Brain Pathol 25:121–135. Shi ZH, Nie G, Duan XL, Rouault T, Wu WS, Ning B (2010) Neuroprotective mechanism of mitochondrial ferritin on

6-hydroxydopamine-induced dopaminergic cell damage: implication for neuroprotection in Parkinson’s disease. Antioxid Redox Signal 13:783–796. Snyder AM, Wang X, Patton SM, Arosio P, Levi S, Earley CJ, Allen RP, Connor JR (2009) Mitochondrial ferritin in the substantia nigra in restless legs syndrome. J Neuropathol Exp Neurol 68:1193–1199. Snyder AM, Neely EB, Levi S, Arosio P, Connor JR (2010) Regional and cellular distribution of mitochondrial ferritin in the mouse brain. J Neurosci Res 88:3133–3143. Wang L, Yang H, Zhao S, Sato H, Konishi Y, Beach TG, Abdelalim EM, Bisem NJ, Tooyama I (2011) Expression and localization of mitochondrial ferritin mRNA in Alzheimer’s disease cerebral cortex. PLoS One 6:e22325. Welch KD, Davis TZ, Van Eden ME, Aust SD (2002) Deleterious ironmediated oxidation of biomolecules. Free Radic Biol Med 32:577–583. Wong-Riley MT (1989) Cytochrome oxidase: an endogenous metabolic marker for neuronal activity. Trends Neurosci 12:94–101. Wu WS, Zhao YS, Shi ZH, Chang SY, Nie GJ, Duan XL, Zhao SM, Wu Q, Yang ZL, Zhao BL, Chang YZ (2013) Correction to mitochondrial ferritin attenuates beta-amyloid-induced neurotoxicity: reduction in oxidative damage through the Erk/ P38 mitogen-activated protein kinase pathways. Antioxid Redox Signal 18:158–169. Yang H, Yang M, Guan H, Liu Z, Zhao S, Takeuchi S (2013) Mitochondrial ferritin in neurodegenerative diseases. Neurosci Res 77:1–7. Yang H, Guan H, Yang M, Liu Z, Takeuchi S, Yanagisawa D, Vincent SR, Zhao S, Tooyama I (2015) Upregulation of mitochondrial ferritin by proinflammatory cytokines: implications for a role in Alzheimer’s disease. J Alzheimers Dis 45:797–811. Zanella I, Derosas M, Corrado M, Cocco E, Cavadini P, Biasiotto G, Poli M, Verardi R, Arosio P (2008) The effects of frataxin silencing in HeLa cells are rescued by the expression of human mitochondrial ferritin. Biochim Biophys Acta 1782:90–98. Zecca L, Youdim MB, Riederer P, Connor JR, Crichton RR (2004) Iron, brain aging and neurodegenerative disorders. Nat Rev Neurosci 5:863–873. Zucca FA, Segura-Aguilar J, Ferrari E, Mun˜oz P, Paris I, Sulzer D, Sarna T, Casella L, Zecca L (2015) Interactions of iron, dopamine and neuromelanin pathways in brain aging and Parkinson’s disease. Prog Neurobiol. pii: S0301-0082(15)00101-X.

(Accepted 22 April 2016) (Available online 28 April 2016)