Iron overload induces hypogonadism in male mice via extrahypothalamic mechanisms

Molecular and Cellular Endocrinology xxx (2017) 1e11

Contents lists available at ScienceDirect

Molecular and Cellular Endocrinology journal homepage: www.elsevier.com/locate/mce

Iron overload induces hypogonadism in male mice via extrahypothalamic mechanisms Chiara Macchi a, Liliana Steffani a, Roberto Oleari a, Antonella Lettieri a, Luca Valenti b, Paola Dongiovanni b, Antonio Romero-Ruiz c, Manuel Tena-Sempere c, d, Anna Cariboni a, **, Paolo Magni a, *, Massimiliano Ruscica a  degli Studi di Milano, 20133 Milan, Italy Department of Pharmacological and Biomolecular Sciences, Universita  degli Studi Milano, UO Medicina Interna 1B, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Pathophysiology and Transplantation, Universita 20100 Milan, Italy c Department of Cell Biology, Physiology and Immunology, University of Cordoba, Instituto Maimonides de Investigacion Biom edica de Cordoba (IMIBIC), Hospital Universitario Reina Sofia, 14004 Cordoba, Spain d n, Instituto de Salud Carlos III, 14004 Cordoba, Spain CIBER Fisiopatología de la Obesidad y Nutricio a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 2 February 2017 Received in revised form 8 June 2017 Accepted 17 June 2017 Available online xxx

Introduction: Iron overload leads to multiple organ damage including endocrine organ dysfunctions. Hypogonadism is the most common non-diabetic endocrinopathy in primary and secondary iron overload syndromes. Aim: To explore the molecular determinants of iron overload-induced hypogonadism with specific focus on hypothalamic derangements. A dysmetabolic male murine model fed iron-enriched diet (IED) and cell-based models of gonadotropin-releasing hormone (GnRH) neurons were used. Results: Mice fed IED showed severe hypogonadism with a significant reduction of serum levels of testosterone (
83%) and of luteinizing hormone (
86%), as well as reduced body weight gain, body fat and plasma leptin. IED mice had a significant increment in iron concentration in testes and in the pituitary. Even if iron challenge of in vitro neuronal models (GN-11 and GT1-7 GnRH cells) resulted in 10and 5-fold iron content increments, respectively, no iron content changes were found in vivo in hypothalamus of IED mice. Conversely, mice placed on IED showed a significant increment in hypothalamic GnRH gene expression (þ34%) and in the intensity of GnRH-neuron innervation of the median eminence (þ1.5-fold); similar changes were found in the murine model HFE
/
, resembling human hemochromatosis. Conclusions: IED-fed adult male mice show severe impairment of hypothalamus-pituitary-gonadal axis without a relevant contribution of the hypothalamic compartment, which thus appears sufficiently protected from systemic iron overload. © 2017 Elsevier B.V. All rights reserved.

Keywords: Hypogonadism Iron overload GnRH Median eminence Testes

1. Introduction Iron is an essential metal for fundamental biochemical activities such as oxygen transport and energy metabolism. It is also required for proper brain development during embryogenesis, and for brain

* Corresponding author. Department of Pharmacological and Biomolecular Sci degli Studi di Milano, via G. Balzaretti, 9, 20133 Milan, Italy. ences, Universita ** Corresponding author. Department of Pharmacological and Biomolecular Sci degli Studi di Milano, via G. Balzaretti, 9, 20133 Milan, Italy. ences, Universita E-mail addresses: [email protected] (A. Cariboni), [email protected] (P. Magni).

function in the early neonatal period and adult age (Radlowski and Johnson, 2013). On the other hand, iron overload is prone to trigger Fenton chemistry resulting in cellular oxidative stress and generation of highly reactive radicals (Chevion, 1988). Pathological iron overload conditions have been associated with dysfunctions of endocrine organs (Brissot et al., 2016). In individuals with primary (hereditary hemochromatosis) and secondary (transfusional and dietary) iron overload syndromes, iron plays a relevant causative role in several clinical manifestations, including diabetes mellitus, insulin resistance (IR) and nonalcoholic fatty liver disease (Datz et al., 2013; Simcox and McClain, 2013). The dysmetabolic iron overload syndrome (DIOS),

http://dx.doi.org/10.1016/j.mce.2017.06.019 0303-7207/© 2017 Elsevier B.V. All rights reserved.

Please cite this article in press as: Macchi, C., et al., Iron overload induces hypogonadism in male mice via extrahypothalamic mechanisms, Molecular and Cellular Endocrinology (2017), http://dx.doi.org/10.1016/j.mce.2017.06.019

2

C. Macchi et al. / Molecular and Cellular Endocrinology xxx (2017) 1e11

also referred to as IR-associated hepatic iron overload (Mendler et al., 1999), corresponds to mild hepatic iron excess in the context of various features of metabolic syndrome (Dongiovanni et al., 2011). Notably, the pathogenesis of iron accumulation in DIOS has been related to altered iron trafficking associated with steatosis, hepatic inflammation, and IR (Dongiovanni et al., 2013). Hypogonadism is the most common non-diabetic endocrinopathy both in primary (McDermott and Walsh, 2005) and secondary (Al-Rimawi et al., 2005) iron overload syndromes. Indeed, the clinical picture of hereditary hemochromatosis is associated to impotence in males (Niederau et al., 1994). Hypogonadism in men results from failure of the testes to produce physiological levels of testosterone (T) with a normal number of spermatozoa due to disruption of one or more levels of the hypothalamic-pituitarytesticular axis (Bhasin et al., 2006). Hypogonadal subjects are routinely classified into those with primary hypogonadism (testicular failure) characterized by low serum T and elevated gonadotropins (luteinizing hormone -LH-, follicle-stimulating hormone -FSH-) or those with secondary hypogonadism (hypothalamic-pituitary failure) with low serum T and low or normal gonadotropins (Tajar et al., 2010). Hypogonadism related to hemochromatosis has been demonstrated to have a central origin, although there is unclear evidence of the specific contribution of hypothalamus vs. pituitary impairment (Duranteau et al., 1993; Piperno et al., 1992). The same mechanism has also been proposed in thalassaemic patients in whom the hypothalamus-pituitary axis is affected by iron in a dosedependent manner (Chatterjee and Katz, 2000). Conversely, consistent with the effect of human chorionic gonadotropin (hCG) administration, testicular function does not seem to be affected by iron overload (Duranteau et al., 1993; Piperno et al., 1992). Due to a lack of evidence clarifying whether the major driver of iron-induced hypogonadism is hypothalamus or pituitary, by using a murine model of dietary iron overload induced by iron-enriched diet (IED) and suitable cell models of GnRH neurons, the present study aimed at dissecting the molecular basis and the neuroendocrine involvement of iron overload-induced hypogonadism, with a specific focus on hypothalamic derangements.

2. Materials and methods 2.1. Animals Five-week-old male C57BL/6J mice were purchased from Charles River Laboratories (Calco, Italy), housed at constant room temperature (RT, 23  C), under 12-h light/dark cycles, with ad libitum access to tap water and food, in compliance with the European Union guidelines. The investigation conforms to the European Commission Directive 2010/63/EU. Animals were fed either standard iron concentration diet containing 180 mg/kg (control group, CTR; n ¼ 15) or an iron-enriched diet with 3% carbonyl-iron (IED group; n ¼ 15) for 11 weeks (wks) (Dongiovanni et al., 2013). Body weight and food ingestion were monitored weekly. At the end of treatment, animals were sacrificed by decapitation under anesthesia (isoflurane/oxygen mixture) between 10:00 a.m. and 12:00 p.m. to avoid circadian variations. Trunk blood was collected and serum, separated by centrifugation, was stored at
20  C until assayed. Hypothalamus, pituitary and testes were dissected and either flash frozen in liquid nitrogen for RNA extraction or fixed in 4% paraformaldeide for immunohistochemistry procedures. Testicular weight and long diameter and perigonadal fat pad weight were measured before freezing or fixation. Hypothalamus from HFE
/
mice, a model of human genetic hemochromatosis (Zhou et al., 1998), has been used as a positive control. 2.2. Reproductive hormone assays LH and T levels were evaluated by radioimmunoassay (RIA) (Pineda et al., 2010). Serum LH levels were determined in a volume of 50 mL using a double-antibody method and RIA kits supplied by the National Institutes of Health (Dr. A. F. Parlow, National Institute of Diabetes and Digestive and Kidney Diseases National Hormone and Peptide Program, Torrance, CA). Rat LH-I-10 was labelled with 125I using Iodogen tubes, following the instructions of the manufacturer (Pierce, Rockford, IL). Hormone concentrations were expressed using reference preparation, LH-RP-3 as standard. Intraand inter-assay coefficients of variation were less than 8 and 10%. The sensitivity of the assay was 5 pg/tube for LH. All samples were

Table 1 Primer sequences. Gene

CHOP FSHb FtH GnRH Gpr54 Kiss1 LHb SOD2 TfR XBP-1 18S

Primers Forward primer (50 -30 )

Reverse primer (50 -30 )

GTCCCTAGCTTGGCTGACAGA ATGGATTGTTCCAGGCAGAC CGAGATGATGTGGCTCTGAA GGCCGGCATTCTACTGCTG CAGTCCCAGGACACAATCCT AGCTGCTGCTTCTCCTCTGT TGGCCGCAGAGAATGAGTTC TCTGGCCAAGGGAGATGTTA TCGCTTATATTGGGCAGACC TGAGAACCAGGAGTTAAGAACACGC CTCGCTCCTCTCCTACTTGG

TGGAGAGCGAGGGCTTTG TCACTGCATGTGAGGGAAAG GTGCACACTCCATTGCATTC CTGCCTGGCTTCCTCTTCA ACCAATGAGTTTCCGACCAG GCATACCGCGATTCCTTTT CTCGGACCATGCTAGGACAGTAG CCTCCAGCAACTCTCCTTTG CCATGTTTTGACCAATGCTG TTCTGGGTAGACCTCTGGGAGTTCC CCATCGAAAGTTGATAGGGC

Applied Biosystems® Custom TaqMan® 5′ FAM e 3′ MGB Probes Assay ID GnRH1 IL-6 Kiss1 NPY 18S

Mm01315605_m1 Mm00446190_m1 Mm03058560_m1 Mm03048253_m1 Hs99999901_s1

FAM: 6-carboxy-fluorescein (reporter fluorescent dye). MGB: minor-groove-binder moiety.

Please cite this article in press as: Macchi, C., et al., Iron overload induces hypogonadism in male mice via extrahypothalamic mechanisms, Molecular and Cellular Endocrinology (2017), http://dx.doi.org/10.1016/j.mce.2017.06.019

C. Macchi et al. / Molecular and Cellular Endocrinology xxx (2017) 1e11

3

measured in the same assay; accuracy of hormone determinations was confirmed by assessment of rat serum samples of known hormone concentrations. Serum T levels were measured using a commercial RIA kit from MP Biomedicals (Costa Mesa, CA). All medium samples were measured in the same assay. The sensitivity of the assay was 1 ng/ mL, and the intra-assay coefficient of variation was 4.5%.

2.5. Intracellular iron content

2.3. Cell cultures

Cell viability was assessed by using the ATPlite™ 1step assay kit (Perkin-Elmer, Monza, Italy) and the trypan-blue exclusion test performed with the automated cell counter Luna™ (Logos Biosystems, Inc., Annandale, VA). Briefly, for the ATPlite assay, GN11 cells were seeded in a 96-well-microplate at 0.02  106 cells/ well and then exposed for 3 or 24 h to 200 mM FAC. ATPlite 1step reagent, containing luciferase and D-luciferin, was added to each well and the luminescence, proportional to ATP concentration, was measured. For the trypan-blue exclusion test, GN-11 cells were seeded in 6-well plates at 0.3  106 cells/well and exposed for 3 or 24 h to 200 mM FAC. The viability was evaluated by mixing 10 mL of the harvested cells with 10 mL of trypan-blue stain.

Mouse GN-11 (a kind gift of Dr S. Radovick, Boston, MA) and GT1-7 cells (a kind gift of Dr R. I. Weiner, San Francisco, CA), representative of immature immortalized and mature GnRH neurons, were grown at 37  C in a humidified CO2 incubator in monolayer. The culture medium was Dulbecco's Minimal Essential Medium (DMEM; Sigma-Aldrich, Milan, Italy) supplemented with 0.0159 g/L phenol red, 4.5 g/L D-glucose, 1 mM sodium pyruvate (Biochrom, Berlin, Germany), 100 mg/mL streptomycin, 100 U/mL penicillin (Sigma-Aldrich), 2 mM L-glutamine and 10% fetal bovine serum (FBS; Life Technologies, Monza, Italy). Complete medium was replaced at 3-day intervals. Confluent cells were harvested with 0.05% trypsin/0.02% ethylenediaminetetraacetic acid (EDTA; Sigma-Aldrich) and seeded at 0.1  106 cells per 100-mm Petri dish. 2.4. Chemicals Ferric ammonium citrate (FAC) was purchased from SigmaAldrich. Desferal® (deferoxamine mesylate, DFO) was obtained from Biofutura Pharma (Milan, Italy).

Hypothalamic, testicular, GN-11 and GT1-7 cells iron content was measured by atomic absorption spectrometry (AAS) as previously reported (Barry and Sherlock, 1971). 2.6. Cell viability assays

2.7. RNA extraction, RT-PCR and quantitative (q) RT-PCR Total RNA was harvested from pituitary, testes and neuronal cells by the RNeasy Mini kit (Qiagen, Milan, Italy) and from the hypothalamus with the NucleoSpin® RNA/Protein kit (MachereyNagel, Düren, Germany) according to manufacturer's instructions. The amount of RNA was quantified with a biophotometer

Fig. 1. Effect of iron-enriched diet on (i) body weight (panel A), (ii) perigonadal fat pad (panel B), (iii) serum leptin levels (panel C), and (iv) hypothalamic neuropeptide Y (NPY) gene expression (panel D). Mice were fed chow-diet (control - CTR - group) or iron-enriched diet (IED group; 3% carbonyl-iron), respectively, for 11 weeks. White circles are controls; black squares are IED mice. Fifteen mice per group were analyzed. In panels A, C and D, data are expressed as mean ± SEM. In panel B, data are expressed as percentage of control (mean ± SEM). Differences between treatments were assessed by Student's t-test, *p < 0.05, **p < 0.01, ***p < 0.001 and ****p < 0.0001. Results are representative of three independent experiments.

Please cite this article in press as: Macchi, C., et al., Iron overload induces hypogonadism in male mice via extrahypothalamic mechanisms, Molecular and Cellular Endocrinology (2017), http://dx.doi.org/10.1016/j.mce.2017.06.019

4

C. Macchi et al. / Molecular and Cellular Endocrinology xxx (2017) 1e11

Fig. 2. Effect of iron-enriched diet (IED) on reproductive function. Testes weight, morphology and the mean number of seminiferous tubules are reported in panels AeC. Serum testosterone (T) and luteinizing hormone (LH) levels in mice fed IED compared to mice fed chow diet are shown in panels DeE. White bars are CTR mice and the black ones are IED mice. Data reported in panels AeC are representative of 5 mice whereas that of panels D and E were 15. Data are expressed as mean ± SEM. Differences between treatments were assessed by Student's t-test, *p < 0.05, **p < 0.01 and ***p < 0.001. Results are representative of three independent experiments.

(Eppendorf, Milan, Italy). One mg of total RNA was retro-transcribed into first-strand cDNA in 20 mL final volume using the iScript Reverse Transcription Supermix (Bio-Rad Laboratories, Segrate, MI, Italy) at the following conditions: 25 C/5 min, 42 C/30 min and 85 C/5 min. Negative control reactions omitting RNA or reverse transcriptase (-RT) were performed in parallel to check for sample or genomic contamination, respectively. To determine X-box binding protein 1 (XBP-1) mRNA splicing, final-time PCR was performed using 50 ng cDNA as a template, specific oligonucleotides (Sigma-Aldrich) and the GoTaq® Green Master Mix (Promega, Milan, Italy) in 25 mL final volume. Amplification products were separated by 4% agarose gel electrophoresis and detected by ethidium bromide fluorescence on a UV transilluminator (Bio-Rad Laboratories). qPCR was carried out by using the SYBR Green Supermix (Bio-Rad Laboratories) in 10 mL final volume or the SsoFast™ Probes Supermix (Bio-Rad Laboratories) in 20 mL final volume with the CFX96 C1000™ Touch system (Bio-Rad Laboratories). For each primer set or probe, samples, no-template controls (no cDNA) and -RT controls were analyzed in triplicate. Sequences of mouse primers (Sigma-Aldrich) and probes (Applied Biosystems® Custom TaqMan® 50 FAM e 30 MGB Probes; Life Technologies) are listed in Table 1. The ribosomal protein 18S was used as internal control. Relative differences in target mRNA levels between control and treated conditions were calculated with the DDCt method.

2.8. ELISA assay Blood serum was collected and stored at
80  C until ELISA analysis. Leptin concentrations were measured by using a Mouse/ Rat Quantikine ELISA kit (R&D systems; Space Import - Export Srl, Milan, Italy). The assay range was 62.5e4,000 pg/mL with a sensitivity of 22 pg/mL.

2.9. Histological analysis and immunostaining Testes and brains were fixed in 4% paraformaldehyde overnight at 4  C. Testes were dehydrated and wax-embedded for a routine H&E staining. Brains were cryopreserved in 30% sucrose and OCTembedded for immunostaining. Formaldehyde-fixed tissue sections were incubated with phosphate-buffered saline (PBS) containing 10% normal goat serum and 0.1% Triton X-100 or, for primary goat antibodies, with serum-free protein block (DAKO). For immunoperoxidase labelling of ME, cryostat sections of formaldehyde-fixed samples were incubated with hydrogen peroxide to quench endogenous peroxidase activity before incubation with GnRH primary antibody (1:1,000), followed by biotinylated goat anti-rabbit antibody (1:400; Vector Laboratories), and then developed with the ABC kit (Vector Laboratories) and DAB (Sigma-Aldrich) (Cariboni et al., 2015).

Please cite this article in press as: Macchi, C., et al., Iron overload induces hypogonadism in male mice via extrahypothalamic mechanisms, Molecular and Cellular Endocrinology (2017), http://dx.doi.org/10.1016/j.mce.2017.06.019

C. Macchi et al. / Molecular and Cellular Endocrinology xxx (2017) 1e11

B

TESTES

PITUITARY

200

12.5

*

**

10.0

mRNA levels (% of CTR)

Iron content (μg/100 mg dry ssue)

A

5

7.5 5.0 2.5 0.0

CTR

150 100 50 0

IED

CTR

IED

FtH

HYPOTHALAMUS

10.0

150

mRNA levels (% of CTR)

7.5 5.0 2.5 0.0 CTR

IED

E 150

mRNA levels (% of CTR)

D

Iron content (μg/100 mg dry ssue)

C

100

50

0 CTR

IED

100

50

0

TfR

CTR

IED

FtH

Fig. 3. Evaluation of iron content and homeostasis in testes, pituitary and hypothalamus. Panel A and C show testes and hypothalamic iron content evaluated by atomic absorption spectrometry. Panel B reports pituitary gene expression of ferritin H (FtH). Panels D and E show hypothalamic transferrin receptor (TfR) and FtH mRNA levels, respectively. White bars are control (CTR) mice and the black ones are iron-enriched diet (IED) mice. In panels A and C data are expressed as mean ± SEM. In panels B, D and E, data are expressed as percentage of control (mean ± SEM). Differences between treatments were assessed by Student's t-test, *p < 0.05 and **p < 0.01. Results are representative of three independent experiments.

2.10. Image processing and quantitation To acquire bright-field images, we used a Zeiss Axiovert 200 with a Photometrics CoolSNAP ES camera (survival assays) or a Leica DM microscope with a DC500 digital camera (AP assays, HRPstained sections). We used a Leica TCS SPE1 confocal microscope to acquire fluorescence images of mouse tissues. All images were processed using Photoshop CS4 (Adobe). To compare the abundance of GnRH-positive neurites at the medial eminence of adult mice fed chow diet or IED and of HFE
/
mice, we measured the pixel intensity of GnRH staining in 20-mm sagittal sections through the ME of 3 mice for each group (control, IED, HFE
/
). For immunoblotting, 3 independent experiments were performed for each condition, the OD of the signal measured with ImageJ software (NIH), and the mean pixel intensity calculated. 2.11. GN-11 cell migration study Microchemotaxis assay was performed with a 48-well Boyden's chamber (Neuro Probe Inc, Gaithersburg, MD) according to the protocol provided. GN-11 cells were treated for 24 h with 200 mM FAC, 200 mM DFO or both. Afterwards, cells were harvested and cell suspension (0.10  106 cells/50 mL DMEM/0.1% bovine serum albumin) was placed in the open-bottom wells of

the upper compartment of the chamber. Each pair of wells was separated by a polyvinyl-pyrrolidone-free polycarbonate porous membrane (8 mm pores; Neuro Probe) pre-coated with gelatin (0.2 mg/mL in PBS; Sigma-Aldrich). 28 mL of 1% fetal bovine serum (FBS) were placed into the lower compartment of the chamber as chemoattractant. After 1.5 h in the cell incubator, the membrane was recovered and cells migrated through the pores were fixed in cold methanol and stained using the Diff-Quick kit (Biomap, Milan, Italy). For quantitative analysis, stained cells were observed using a 20X objective on a light microscope and six random objective fields per well were counted. The mean number of migrated cells/mm2 was expressed as percentage of positive control (1% FBS was taken as 100%).

2.12. Analysis of the data Statistical analysis was performed using the Prism statistical analysis package (GraphPad Software, San Diego, CA). Data are expressed as mean ± standard error of the mean (SEM). Differences between treatment groups were evaluated by t-test or ANOVA, followed by post-hoc Dunnett's or Tukey's test and considered significant at p < 0.05.

Please cite this article in press as: Macchi, C., et al., Iron overload induces hypogonadism in male mice via extrahypothalamic mechanisms, Molecular and Cellular Endocrinology (2017), http://dx.doi.org/10.1016/j.mce.2017.06.019

6

C. Macchi et al. / Molecular and Cellular Endocrinology xxx (2017) 1e11

Fig. 4. Effect of iron on gonadotropin release hormone (GnRH) neurons. In panel A, the hypothalamic gene expression of gonadotropin release hormone (GnRH), kisspeptin and kisspeptin receptor (GPR54) are reported. Panel B shows GnRH neurites staining in the median eminence of mice fed standard and IED diet and HFE
/
mice, a model of human hemochromatosis. Pictures are referred to sagittal sections of 20-mm. Panel C reports the GnRH quantification expressed as pixel intensity. Panels D and E report the hypothalamic gene expression of oxidative and endoplasmic reticulum stress (superoxide dismutase 2 - SOD2 and CAAT/enhancer binding protein (C/EBP) homologous protein - CHOP, respectively) as well as that of X-box binding protein-1 (XBP-1), an index of early stage endoplasmic reticulum stress response. Data in panels DeE were obtained by five different mice belonging to each group and expressed as percentage of control (mean ± SEM). Differences between treatments were assessed by Student's t-test, **p < 0.01. Scale bar, 50 mm in B.

3. Results 3.1. Phenotypic effects of IED IED was associated with a significant reduction in weekly food intake (17.9 ± 0.98 g per mouse) if compared to control (CTR) diet (23.8 ± 0.97 g, p < 0.0001), a change reflected in a lower body weight gain. Specifically, body weight differences between the two groups were already significant after 1 week (wk) of IED (
2.5 g vs CTR, p < 0.05) and progressively increased throughout the 11 wks time-frame (
5.41 g vs CTR at 11 wk; Fig. 1A). As a mirror of this status (Fig. 1B), perigonadal fat pad weight in IED mice was significantly reduced by 57% (CTR: 0.80 ± 0.1 g; IED: 0.46 ± 0.04 g). As previously reported (Dongiovanni et al., 2013), we confirmed that serum leptin levels were reduced in mice IED fed (
56% vs CTR, p < 0.05; Fig. 1C); upon IED, the hypothalamic expression of the gene encoding the orexigenic signal, neuropeptide Y (NPY), was significantly increased by 158% vs CTR (p < 0.05; Fig. 1D). 3.2. Effect of IED on iron homeostasis at different levels of the HPG axis An analysis on a subgroup of animals (n ¼ 5) showed that IED led to a reduction in testicular weight (
22% vs CTR, p < 0.01; data not shown) and longer diameter (21% vs CTR, p < 0.01; data not shown). These differences were maintained after adjustment for body weight (testicular weight:
30% vs CTR, p < 0.001; Fig. 2A; testicular long

diameter:
29% vs CTR, p < 0.001; data not shown). In order to investigate the possible existence of a gonadal defect in males, the morphology of the testes of both groups was analyzed. Eosin and hematoxylin staining of sections and count of seminiferous tubules revealed that the number was significantly reduced (
20%, p < 0.01) in mice fed IED compared with controls (Fig. 2BeC). IED was associated to a pronounced reduction of serum levels of T (0.24 ± 0.10 vs 1.39 ± 0.55 ng/mL, IED and CTR, respectively; p ¼ 0.058; Fig. 2D) and of LH (0.25 ± 0.15 vs 1.82 ± 0.43 ng/mL, IED and CTR, respectively; p < 0.01; Fig. 2E). In mice IED-fed, pituitary mRNA levels of LH-specific beta-subunit (LH-b) also showed a trend toward reduction (
25% vs CTR, p ¼ 0.156; data not shown), whereas those of FSH-specific beta-subunit (FSH-b) were unchanged (data not shown). 3.2.1. Testes AAS showed that mice IED-fed had a significant increment (þ47%, p < 0.01) in testes iron concentration with respect to CTR (9.80 ± 0.82 vs 6.57 ± 0.83 mg/100 mg dry tissue, respectively; Fig. 3A). Testes mRNA levels of the transferrin receptor TfR (responsible for iron transport) and the iron storage protein ferritin H (FtH) were unchanged (data not shown). 3.2.2. Pituitary Pituitary iron content was not quantified by AAS, due to methodological limitations, but was estimated by evaluating FtH gene expression, which is considered as a surrogate biomarker of iron

Please cite this article in press as: Macchi, C., et al., Iron overload induces hypogonadism in male mice via extrahypothalamic mechanisms, Molecular and Cellular Endocrinology (2017), http://dx.doi.org/10.1016/j.mce.2017.06.019

C. Macchi et al. / Molecular and Cellular Endocrinology xxx (2017) 1e11

7

GN-11 CTR 200 μM FAC (24h)

CTR 200 μM FAC (24h)

A

B

0.05

150

*

200

mRNA levels (% of CTR)

0.10

C

250

**** mRNA levels (% of CTR)

0.15

Iron content (μg/106 cells)

CTR 200 μM FAC (24h)

150 100

****

50

0

0

100

50

0

TfR

FtH

GnRH

GT1-7 CTR 200 μM FAC (24h)

CTR 200 μM FAC (24h)

D

E

0.05

0.02

250 200 150 100

0.01

50

0.00

0

150

* mRNA levels (% of CTR)

0.03

F

**** mRNA levels (% of CTR)

Iron content (μg/106 cells)

0.04

CTR 200 μM FAC (24h)

100

50

*** 0

TfR

FtH

GnRH

Fig. 5. Effect of iron treatment in GN-11 and GT1-7 cells. Panels AeC show the effect of 24-h treatment with ferric ammonium citrate (FAC; 200 mM) on neuronal immature murine GN-11 cells. Iron content and genes involved in iron homeostasis are reported in panels A and B, respectively. Panel C reports findings related to gonadotropin release hormone (GnRH) gene expression. The above reported parameters were also evaluated on neuronal mature murine GT1-7 cells (panels DeF). White bars are controls (CTR) and the black ones are FAC treated. In panels A and D data are expressed as mean ± SEM. In panels B-C and E-F, data are expressed as percentage of control (mean ± SEM). Differences between treatments were assessed by Student's t-test, *p < 0.05, ***p < 0.001 and ****p < 0.0001. Results are representative of three independent experiments.

content (Li et al., 2013). Pituitary FtH mRNA was significantly increased in IED mice with respect to CTR (þ48%, p < 0.05; Fig. 3B), suggesting iron accumulation. 3.2.3. Hypothalamus At sacrifice, AAS showed no differences in the iron content of the whole hypothalamus between IED and CTR mice (6.37 ± 1.91 vs 6.37 ± 2.28 mg/100 mg dry tissue, respectively; Fig. 3C). As a further confirmation, no differences were found for the gene expression of TfR and FtH (Fig. 3DeE). 3.3. Effect of IED on hypothalamic and pituitary factors controlling reproduction, and oxidative and endoplasmic reticulum stress Mice fed with IED showed a significant increment in hypothalamic GnRH gene expression (þ34%, p < 0.01); conversely Kiss1 and the related G protein-coupled receptor 54 (GPR54), two master key regulators of reproduction (Pinilla et al., 2012), were unchanged (Fig. 4A). To further evaluate the effect of IED on the reproductive axis, we evaluated the presence of GnRH-positive neurites at the level of the ME, where they release GnRH into the portal blood vessels of the pituitary gland. As shown in Fig. 4BeC, the ME of mice fed with IED was much more innervated by GnRH-positive neurites, as evaluated by GnRH pixel intensity increment (1.5-fold) in ME

neurites compared to control diet mice. This is probably due to lack of negative feedback from circulating T. Moreover, compared to mice fed standard diet, a more intense staining of GnRH-positive neurites (1.4-fold) to the ME was also observed in HFE
/
mice, a model of human genetic hemochromatosis (Fig. 4B). Since tissue iron loading is a well-known inducer of oxidative stress and unfolded protein response, we further evaluated the effect of iron on these parameters. No changes were found for the hypothalamic gene expression of (i) superoxide dismutase 2 (SOD2), a well-known marker of oxidative stress (
18% vs CTR, p ¼ 0.09; Fig. 4D), (ii) ER stress-related CAAT/enhancer binding protein (C/EBP) homologous protein (CHOP; p ¼ 0.342 vs CTR; Fig. 4D), and (iii) X-Box Binding Protein-1 (XBP-1) spliced form (corresponding to a 300-bp band), an index of early stage ER stress response (Fig. 4E). Similarly, no changes were seen for hypothalamic interleukin-6 (IL-6) gene expression (data not shown). 3.4. Effect of FAC on iron homeostasis in GN-11 and GT1-7 cells To understand the selective contribution of the Blood Brain Barrier (BBB) on protection of hypothalamus from iron accumulation, GnRH neuronal models (GN-11 and GT1-7 cells) were used. A 24-h treatment of GN-11 and GT1-7 cells with 200 mM FAC increased intracellular iron concentration by 10-fold (p < 0.0001)

Please cite this article in press as: Macchi, C., et al., Iron overload induces hypogonadism in male mice via extrahypothalamic mechanisms, Molecular and Cellular Endocrinology (2017), http://dx.doi.org/10.1016/j.mce.2017.06.019

8

C. Macchi et al. / Molecular and Cellular Endocrinology xxx (2017) 1e11

Fig. 6. Effect of iron treatment on GN-11 cells oxidative stress and migratory capacity. Panels AeB show the effect of 24-h treatment with ferric ammonium citrate (FAC; 200 mM) on gene expression of oxidative and endoplasmic reticulum stress (superoxide dismutase 2 - SOD2 and CAAT/enhancer binding protein (C/EBP) homologous protein - CHOP, respectively) as well as that of X-box binding protein-1 (XBP-1), an index of early stage endoplasmic reticulum stress response. White bars are controls (CTR) and the black ones are FAC treated. Panel C reports chemotactic and chemokinetic response of GN-11 cells to FAC (200 mM) and deferoxamine mesylate (DFO; 200 mM). The assay was performed using a Boyden's chamber assay. In panels A and C, data are expressed as percentage of control (mean ± SEM). Differences between treatments were assessed by Student's t-test (panels AeB) and by ANOVA (panel C), *p < 0.05 and ***p < 0.001. Results are representative of three independent experiments.

and 5-fold (p < 0.0001), respectively, as evaluated by AAS (Fig. 5A,D), without affecting cell viability and morphology (suppl. Fig. 1). The ability of our cell models to adapt to exogenous iron overload was confirmed by the reduction of TfR mRNA levels (GN11:
58% vs CTR, p < 0.0001; GT1-7:
75% vs CTR, p < 0.001) and the increment of FtH mRNA (GN-11: þ83% vs CTR, p < 0.05; GT17: þ92% vs CTR, p < 0.05 Fig. 5B,E). FAC did not affect GnRH gene expression either in GN-11 cells (p ¼ 0.297; Fig. 5C) or in GT1-7 cells (p ¼ 0.429; Fig. 5F). A 24-h FAC treatment up-regulated IL-6 gene expression (þ46% vs CTR, p < 0.05; data not shown), and increased SOD2 mRNA levels (þ51% vs CTR, p < 0.05, Fig. 6A), without affecting those of CHOP and of alternative splicing of XBP-1 (Fig. 6B and C). The Boyden chamber-based microchemotaxis assay was performed on FAC-treated GN-11 cells, which possess migratory properties. We exposed these cells to 1% FBS, a recognized potent chemoattractant, in order to obtain a migratory response (Ruscica

et al., 2011). This condition was assumed as internal positive control, in terms of percentage of migrated cells. A 24-h pretreatment of GN-11 cells with 200 mM FAC, before exposure for 1.5 h to 1% FBS in Boyden's chamber, inhibited 1% FBS-induced chemotaxis by 19% (p < 0.05, Fig. 6C). The negative contribution of FAC on neuronal migration was confirmed by a 24-h cotreatment of GN-11 cells with FAC and DFO, an hexadentate chelator, at a 1:1 M ratio (iron(III)-chelator complexes; 200 mM) before Boyden's chamber assay. This association completely counteracted the inhibition mediated by FAC on 1% FBS-induced GN-11 migration (Fig. 6C). 4. Discussion This study addressed the issue of the possible contribution of hypothalamic derangement to the impairment of the HPG axis induced by dietary iron overload. Indeed, to date it remains

Please cite this article in press as: Macchi, C., et al., Iron overload induces hypogonadism in male mice via extrahypothalamic mechanisms, Molecular and Cellular Endocrinology (2017), http://dx.doi.org/10.1016/j.mce.2017.06.019

C. Macchi et al. / Molecular and Cellular Endocrinology xxx (2017) 1e11

unknown whether iron-driven hypogonadism involves the hypothalamus. To address this issue, we used a murine model of dietary iron overload, which showed a doubling of serum iron content (from 40 to 80 mg/dL) (Dongiovanni et al., 2013). The cumulative evidence collected in the present study suggests that exogenous (dietary) iron overload results in a major impairment of the HPG axis without a relevant involvement of the hypothalamus. However, our in vitro studies, in which the BBB is obviously lacking, suggest some degree of vulnerability of GnRH-secreting neurons. The most common clinical conditions associated with iron overload, i.e., hemochromatosis and b-thalassemia, have been associated to some endocrinopathies, including hypogonadism (Ferro et al., 2017; Pelusi et al., 2016). This syndrome, caused by the disruption of one or more levels of the HPG axis, results from failure of testes to produce physiological levels of T and normal number of spermatozoa (Bhasin et al., 2006). A possible iron accumulation at the hypothalamic-pituitary level may impair GnRH neurons and/or pituitary gonadotroph cells leading to hypogonadotropic hypogonadism, characterized by low levels of gonadotropins (FSH and LH) and T (Dandona and Rosenberg, 2010). In patients with hemochromatosis, subnormal T levels with low or inappropriately normal basal gonadotropin concentrations and a blunted or absent response of FSH and LH to GnRH administration are the main hormonal parameters reported in male hypogonadic patients, thus supporting a pituitary dysfunction (Pelusi et al., 2016). Indeed, these patients show a pituitary unresponsiveness to chronic pulsatile GnRH therapy without a significant increment in the mean levels of LH, FSH and T (Duranteau et al., 1993). This evidence agrees with our model, in which the HPG impairment driven by iron overload included both pituitary and testicular changes, as indicated by decreased serum LH and T levels by 86% and 83%, respectively, and by reduced testicular weight and number of seminiferous tubules. Notably, although in patients with hemochromatosis iron deposition was described in pituitary gonadotroph, somatotroph, lactotroph, corticotroph and thyrotroph cells, accumulation of iron appears to be more pronounced in gonadotrophs than in other cell types (Bergeron and Kovacs, 1978). Recent experimental data show that a single acute intraperitoneal injection of iron-dextran in female rats led to dose-dependent iron accumulation in hypothalamus, pituitary and ovaries, with an apparently main involvement of the latter, without changes of circulating LH and FSH (Rossi et al., 2016). In line with these data, our experimental model (male mice IED-fed) also showed an increment of iron content in testes, evaluated by AAS, and in pituitary, as assessed by the increased gene expression of FtH, a surrogate biomarker of iron content (Li et al., 2013), but not in the hypothalamus, differently from what reported by Rossi et al. (2016). Since we found a higher expression of hypothalamic GnRH in mice fed IED, we further investigated the presence of GnRH neurons projections at the ME, where GnRH is released into the portal blood vessels of the pituitary gland (Silverman et al., 1989). Of note, IED mice showed an increment in GnRH projections to the ME, similar to HFE
/
mice, a model of human hemochromatosis. This evidence suggests that, in our IED murine model, low levels of LH and T are related to primary defects downstream of the hypothalamus. In fact, the increased hypothalamic expression of GnRH (detected at the mRNA and protein levels) is likely to be primarily due to the lack of a selective negative feedback driven by T in IED mice (Crowley, 2011; Fraietta et al., 2013). Thus, our findings seem to rule out the involvement of the hypothalamus in iron-driven hypogonadism, which was instead previously hypothesized, based on normal pituitary responses to GnRH administration in a few patients with hemochromatosis but with mild iron accumulation (Piperno et al., 1992). Nevertheless, although as in our model gonadal atrophy was

9

described in hereditary hemochromatosis (Kelly et al., 1984), our current data cannot rule out the contribution of direct impairment of testicular function to the state of hypogonadism. In this context, previous studies reported how the administration of iron to rats determined its accumulation in the interstitium, interstitial macrophages, and Leydig cells. This was associated with testicular atrophy, morphological changes, impaired spermatogenesis, epididymal lesions, and impaired reproductive performance (Lucesoli et al., 1999). However, in patients with hemochromatosis the accumulation of iron in the testes is absent or rare (Siemons and Mahler, 1987) and a T response to hCG administration similar to healthy subjects was observed (Duranteau et al., 1993). In the context of reproduction, the role of the adipokine leptin, a signal of nutritional state, should also be considered, since it represents a permissive factor for the reproductive axis (Hausman et al., 2012; Magni, 2003; Roa and Tena-Sempere, 2014). Upon IED, we observed a decrement of serum leptin levels and of perigonadal fat pad and an increment of hypothalamic gene expression of the orexigenic peptide NPY (Muroi and Ishii, 2016). While the effects of NPY on GnRH neurosecretion are probably dual (stimulatory and inhibitory) and dependent on the prevailing metabolic and reproductive state, experimental evidence has suggested a major inhibitory action of NPY signaling in the control of key aspects of HPG axis function, such as puberty onset and LH secretion (Catzeflis et al., 1993; Pralong, 2010). Further investigations are necessary to understand whether the observed increase in hypothalamic NPY expression plays a role in the suppression of LH levels in IED male mice. Intriguingly, we did not find any iron accumulation at hypothalamic level. Indeed, the gene expression of TfR, the receptor of plasma iron carrier transferrin, and of the main iron storage protein FtH (Ganz, 2013) was unchanged between controls and IED. A possible explanation is that the hypothalamus may be protected from systemic iron overload by the BBB. Indeed, when in vitro neuronal models (GN-11 and GT1-7 immortalised GnRH cells), devoid of BBB, were treated with exogenous iron, massive

Fig. 7. This figure summarizes the main findings of the study. In adult male mice, dietary iron overload leads to a severe impairment of hypothalamus-pituitary-gonadal axis without a relevant contribution of the hypothalamic compartment. FSH, folliclestimulating hormone; FtH, ferritin H; GnRH, gonadotropin-releasing hormone; LH, luteinizing hormone; T, testosterone.

Please cite this article in press as: Macchi, C., et al., Iron overload induces hypogonadism in male mice via extrahypothalamic mechanisms, Molecular and Cellular Endocrinology (2017), http://dx.doi.org/10.1016/j.mce.2017.06.019

10

C. Macchi et al. / Molecular and Cellular Endocrinology xxx (2017) 1e11

iron accumulation was observed, in association with impaired iron metabolism and upregulation of the SOD2 gene. Conversely, in animals fed IED, hypothalamic SOD2, C/EBP and XBP-1 gene expression was coherently not affected. In conclusion, using a translational model of iron overload, we have documented that such condition is associated with a marked state of hypogonadism, which closely resembles clinical states of iron excess. However, the hypothalamic compartment seems to be sufficiently protected from systemic dietary iron overload in adult male mice and does not appear to remarkably contribute to the massive HPG axis suppression produced by IED (Fig. 7). Moreover, based on data reporting how phlebotomy recovered both pituitary and gonadal function, especially when this intervention is introduced at the early phases of disease (Pelusi et al., 2016), future studies addressing the use of chelation therapy can be considered to restore an eugonadic status. Indeed, thalassaemic patients, with iron accumulation due to transfusion therapy, treated for at least 5 years with the iron chelator, deferasirox, showed a decreased prevalence of hypogonadism (Poggi et al., 2016). Disclosure All the authors have nothing to disclose. Conflict of interest All the authors have nothing to disclose. Acknowledgments Cariplo Foundation (Ref. 2015-0552) and intramural grant by  degli Studi di Milano (linea 2 e Azione A 2015e2017) to Universita MR. The financial support of Telethon - Italy (Grant no. 13142 to AC) is gratefully acknowledged. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.mce.2017.06.019. References Al-Rimawi, H.S., Jallad, M.F., Amarin, Z.O., Obeidat, B.R., 2005. Hypothalamic-pituitary-gonadal function in adolescent females with beta-thalassemia major. Int. J. Gynaecol. Obstet. 90, 44e47. Barry, M., Sherlock, S., 1971. Measurement of liver-iron concentration in needlebiopsy specimens. Lancet 1, 100e103. Bergeron, C., Kovacs, K., 1978. Pituitary siderosis. A histologic, immunocytologic, and ultrastructural study. Am. J. Pathol. 93, 295e309. Bhasin, S., Cunningham, G.R., Hayes, F.J., Matsumoto, A.M., Snyder, P.J., Swerdloff, R.S., Montori, V.M., 2006. Testosterone therapy in adult men with androgen deficiency syndromes: an endocrine society clinical practice guideline. J. Clin. Endocrinol. Metab. 91, 1995e2010. Brissot, P., Cavey, T., Ropert, M., Gaboriau, F., Loreal, O., 2016. Hemochromatosis: a model of metal-related human toxicosis. Environ. Sci. Pollut. Res. Int. http:// dx.doi.org/10.1007/s11356-016-7576-2. First Online: 15 September 2016. Cariboni, A., Andre, V., Chauvet, S., Cassatella, D., Davidson, K., Caramello, A., Fantin, A., Bouloux, P., Mann, F., Ruhrberg, C., 2015. Dysfunctional SEMA3E signaling underlies gonadotropin-releasing hormone neuron deficiency in Kallmann syndrome. J. Clin. Invest. 125, 2413e2428. Catzeflis, C., Pierroz, D.D., Rohner-Jeanrenaud, F., Rivier, J.E., Sizonenko, P.C., Aubert, M.L., 1993. Neuropeptide Y administered chronically into the lateral ventricle profoundly inhibits both the gonadotropic and the somatotropic axis in intact adult female rats. Endocrinology 132, 224e234. Chatterjee, R., Katz, M., 2000. Reversible hypogonadotrophic hypogonadism in sexually infantile male thalassaemic patients with transfusional iron overload. Clin. Endocrinol. (Oxf) 53, 33e42. Chevion, M., 1988. A site-specific mechanism for free radical induced biological damage: the essential role of redox-active transition metals. Free Radic. Biol. Med. 5, 27e37. Crowley, W.F., 2011. The developmental biology of the GnRH neurons. Mol. Cell Endocrinol. 346, 1e3.

Dandona, P., Rosenberg, M.T., 2010. A practical guide to male hypogonadism in the primary care setting. Int. J. Clin. Pract. 64, 682e696. Datz, C., Felder, T.K., Niederseer, D., Aigner, E., 2013. Iron homeostasis in the metabolic syndrome. Eur. J. Clin. Invest. 43, 215e224. Dongiovanni, P., Fracanzani, A.L., Fargion, S., Valenti, L., 2011. Iron in fatty liver and in the metabolic syndrome: a promising therapeutic target. J. Hepatol. 55, 920e932. Dongiovanni, P., Ruscica, M., Rametta, R., Recalcati, S., Steffani, L., Gatti, S., Girelli, D., Cairo, G., Magni, P., Fargion, S., Valenti, L., 2013. Dietary iron overload induces visceral adipose tissue insulin resistance. Am. J. Pathol. 182, 2254e2263. Duranteau, L., Chanson, P., Blumberg-Tick, J., Thomas, G., Brailly, S., Lubetzki, J., Schaison, G., Bouchard, P., 1993. Non-responsiveness of serum gonadotropins and testosterone to pulsatile GnRH in hemochromatosis suggesting a pituitary defect. Acta Endocrinol. (Copenh) 128, 351e354. Ferro, E., Visalli, G., La Rosa, M.A., Piraino, B., Civa, R., Randazzo Papa, G., Di Pietro, A., 2017. Genotoxic effect of iron overload and disease complications in transfused beta thalassaemic patients. Mutagenesis 32, 275e281. Fraietta, R., Zylberstejn, D.S., Esteves, S.C., 2013. Hypogonadotropic hypogonadism revisited. Clin. (Sao Paulo) 68 (Suppl. 1), 81e88. Ganz, T., 2013. Systemic iron homeostasis. Physiol. Rev. 93, 1721e1741. Hausman, G.J., Barb, C.R., Lents, C.A., 2012. Leptin and reproductive function. Biochimie 94, 2075e2081. Kelly, T.M., Edwards, C.Q., Meikle, A.W., Kushner, J.P., 1984. Hypogonadism in hemochromatosis: reversal with iron depletion. Ann. Intern Med. 101, 629e632. Li, J., Wang, R., Luo, D., Li, S., Xiao, C., 2013. Association between serum ferritin levels and risk of the metabolic syndrome in Chinese adults: a population study. PLoS One 8, e74168. Lucesoli, F., Caligiuri, M., Roberti, M.F., Perazzo, J.C., Fraga, C.G., 1999. Dose-dependent increase of oxidative damage in the testes of rats subjected to acute iron overload. Arch. Biochem. Biophys. 372, 37e43. Magni, P., 2003. Hormonal control of the neuropeptide Y system. Curr. Protein Pept. Sci. 4, 45e57. McDermott, J.H., Walsh, C.H., 2005. Hypogonadism in hereditary hemochromatosis. J. Clin. Endocrinol. Metab. 90, 2451e2455. Mendler, M.H., Turlin, B., Moirand, R., Jouanolle, A.M., Sapey, T., Guyader, D., Le Gall, J.Y., Brissot, P., David, V., Deugnier, Y., 1999. Insulin resistance-associated hepatic iron overload. Gastroenterology 117, 1155e1163. Muroi, Y., Ishii, T., 2016. A novel neuropeptide Y neuronal pathway linking energy state and reproductive behavior. Neuropeptides 59, 1e8. Niederau, C., Strohmeyer, G., Stremmel, W., 1994. Epidemiology, clinical spectrum and prognosis of hemochromatosis. Adv. Exp. Med. Biol. 356, 293e302. Pineda, R., Garcia-Galiano, D., Sanchez-Garrido, M.A., Romero, M., Ruiz-Pino, F., Aguilar, E., Dijcks, F.A., Blomenrohr, M., Pinilla, L., van Noort, P.I., TenaSempere, M., 2010. Characterization of the potent gonadotropin-releasing activity of RF9, a selective antagonist of RF-amide-related peptides and neuropeptide FF receptors: physiological and pharmacological implications. Endocrinology 151, 1902e1913. Pelusi, C., Gasparini, D.I., Bianchi, N., Pasquali, R., 2016. Endocrine dysfunction in hereditary hemochromatosis. J. Endocrinol. Invest. 39, 837e847. Pinilla, L., Aguilar, E., Dieguez, C., Millar, R.P., Tena-Sempere, M., 2012. Kisspeptins and reproduction: physiological roles and regulatory mechanisms. Physiol. Rev. 92, 1235e1316. Piperno, A., Rivolta, M.R., D'Alba, R., Fargion, S., Rovelli, F., Ghezzi, A., Micheli, M., Fiorelli, G., 1992. Preclinical hypogonadism in genetic hemochromatosis in the early stage of the disease: evidence of hypothalamic dysfunction. J. Endocrinol. Invest. 15, 423e428. Poggi, M., Sorrentino, F., Pugliese, P., Smacchia, M.P., Daniele, C., Equitani, F., Terlizzi, F., Guitarrini, M.R., Monti, S., Maffei, L., Losardo, A., Pasin, M., Toscano, V., 2016. Longitudinal changes of endocrine and bone disease in adults with beta-thalassemia major receiving different iron chelators over 5 years. Ann. Hematol. 95, 757e763. Pralong, F.P., 2010. Insulin and NPY pathways and the control of GnRH function and puberty onset. Mol. Cell Endocrinol. 324, 82e86. Radlowski, E.C., Johnson, R.W., 2013. Perinatal iron deficiency and neurocognitive development. Front. Hum. Neurosci. 7, 585. Roa, J., Tena-Sempere, M., 2014. Connecting metabolism and reproduction: roles of central energy sensors and key molecular mediators. Mol. Cell Endocrinol. 397, 4e14. Rossi, E.M., Marques, V.B., Nunes Dde, O., Carneiro, M.T., Podratz, P.L., Merlo, E., dos Santos, L., Graceli, J.B., 2016. Acute iron overload leads to hypothalamicpituitary-gonadal axis abnormalities in female rats. Toxicol. Lett. 240, 196e213. Ruscica, M., Dozio, E., Steffani, L., Passafaro, L., Mazzer, M., Castellano, J.M., Motta, M., Tena-Sempere, M., Magni, P., 2011. Role of the energy sensor adenosine monophosphate-activated protein kinase in the regulation of immature gonadotropin-releasing hormone neuron migration. J. Endocrinol. Invest. 34, e362ee368. Siemons, L.J., Mahler, C.H., 1987. Hypogonadotropic hypogonadism in hemochromatosis: recovery of reproductive function after iron depletion. J. Clin. Endocrinol. Metab. 65, 585e587. Silverman, R.C., Silverman, A.J., Gibson, M.J., 1989. Identification of gonadotropin releasing hormone (GnRH) neurons projecting to the median eminence from third ventricular preoptic area grafts in hypogonadal mice. Brain Res. 501, 260e268. Simcox, J.A., McClain, D.A., 2013. Iron and diabetes risk. Cell Metab. 17, 329e341. Tajar, A., Forti, G., O'Neill, T.W., Lee, D.M., Silman, A.J., Finn, J.D., Bartfai, G.,

Please cite this article in press as: Macchi, C., et al., Iron overload induces hypogonadism in male mice via extrahypothalamic mechanisms, Molecular and Cellular Endocrinology (2017), http://dx.doi.org/10.1016/j.mce.2017.06.019

C. Macchi et al. / Molecular and Cellular Endocrinology xxx (2017) 1e11 Boonen, S., Casanueva, F.F., Giwercman, A., Han, T.S., Kula, K., Labrie, F., Lean, M.E., Pendleton, N., Punab, M., Vanderschueren, D., Huhtaniemi, I.T., Wu, F.C., Group, E., 2010. Characteristics of secondary, primary, and compensated hypogonadism in aging men: evidence from the European male ageing study. J. Clin. Endocrinol. Metab. 95, 1810e1818.

11

Zhou, X.Y., Tomatsu, S., Fleming, R.E., Parkkila, S., Waheed, A., Jiang, J., Fei, Y., Brunt, E.M., Ruddy, D.A., Prass, C.E., Schatzman, R.C., O'Neill, R., Britton, R.S., Bacon, B.R., Sly, W.S., 1998. HFE gene knockout produces mouse model of hereditary hemochromatosis. Proc. Natl. Acad. Sci. U. S. A. 95, 2492e2497.

Please cite this article in press as: Macchi, C., et al., Iron overload induces hypogonadism in male mice via extrahypothalamic mechanisms, Molecular and Cellular Endocrinology (2017), http://dx.doi.org/10.1016/j.mce.2017.06.019