Journal of Chemical Neuroanatomy 61–62 (2014) 64–71

Contents lists available at ScienceDirect

Journal of Chemical Neuroanatomy journal homepage: www.elsevier.com/locate/jchemneu

Differential distribution of hypoxia-inducible factor 1-beta (ARNT or ARNT2) in mouse substantia nigra and ventral tegmental area J.A.D. Dela Cruz a,b, R. Schmidt-Kastner a,c, J.A.A. Stevens a, H.W.M. Steinbusch a, B.P.F. Rutten a,* a School of Mental Health and Neuroscience (MHeNs), Faculty of Health, Medicine and Life Sciences, Maastricht University Medical Center, Maastricht, Netherlands b Department of Psychology, Neuropsychology Doctoral Subprogram, The City University of New York (CUNY) Graduate Center, New York, NY, USA c C.E. Schmidt College of Medicine, Florida Atlantic University (FAU), Boca Raton, FL, USA

A R T I C L E I N F O

A B S T R A C T

Article history: Received 11 February 2014 Received in revised form 3 June 2014 Accepted 2 July 2014 Available online 10 July 2014

Hypoxia has been proposed as a mechanism underlying gene–environment interactions in the neurodevelopmental model of schizophrenia, and hypoxia-inducible factor 1 (HIF-1) could mediate the interactions. In the current study, we analyzed the HIF-1 beta subunit, as formed by aryl hydrocarbon receptor nuclear translocator (ARNT) or ARNT2, in the mouse substantia nigra (SN) and the ventral tegmental area (VTA). We performed immunohistochemical studies of ARNT and ARNT2 in the adult mouse brain, and colocalization analyses, with specific emphasis on dopaminergic cells, i.e. tyrosine hydroxylase (TH) immunoreactive cells. Bioinformatic analyses identified shared protein partners for ARNT and ARNT2. ARNT immunoreactivity showed widespread neuronal labeling without overt regional specificity. We observed co-localization of ARNT and TH in the SN compacta and VTA. Nuclei strongly labeled for ARNT2 were observed in the SN reticulata, while only weak immunoreactivity for ARNT2 was found in TH-immunoreactive neurons in SN compacta and VTA. Stereological analysis showed that ARNT was preferentially expressed in dopaminergic neurons in SN compacta and VTA. Nuclei strongly labeled for ARNT2 were present in neocortex and CA1 of hippocampus. Differential expression of ARNT and ARNT2 in dopaminergic neurons may relate to the vulnerability of distinct dopaminergic projections to hypoxia and to functional vulnerability in schizophrenia and other neuropsychiatric disorders. ß 2014 Elsevier B.V. All rights reserved.

Keywords: HIF-1 Hypoxia Tyrosine hydroxylase Dopamine Aryl hydrocarbon receptor nuclear translocator Schizophrenia

Introduction Molecular regulation of hypoxia by the hypoxia-inducible factor 1 (HIF-1) is primarily of interest for acute ischemia-hypoxia of the brain (Kietzmann et al., 2001; Prabhakar and Semenza, 2012; Sharp and Bernaudin, 2004). A role for hypoxia and HIF-1 has been suggested in the context of the etiology of neuropsychiatric disorders. As such, hypoxia has been proposed as common mechanisms underlying gene–environment interactions in the neurodevelopmental model of schizophrenia (SCZ) (SchmidtKastner et al., 2012). Analysis of genomic data (Richards et al., 2012) and exome sequencing (Gulsuner et al., 2013) has shown that HIF-1A is associated with SCZ, while pathway annotations in

* Corresponding author at: School of Mental Health and Neuroscience, Maastricht University, PO Box 616, 6200 MD Maastricht, Netherlands. Tel.: +31 43 3884120; fax: +31 43 3876038. E-mail address: [email protected] (B.P.F. Rutten). http://dx.doi.org/10.1016/j.jchemneu.2014.07.001 0891-0618/ß 2014 Elsevier B.V. All rights reserved.

other genomic studies of SCZ also highlighted hypoxia genes (Gilman et al., 2012). Other recent genetic studies have shown variants in the HIF-1 system to be associated also with autism spectrum disorders (Neale et al., 2012; O’Roak et al., 2012) as well as mood disorders (Shibata et al., 2013). Most studies of HIF-1 in the brain have focused on the hypoxiaregulated HIF-1 alpha subunit (HIF-1a), which has been reported to show generalized neuronal expression (Acker and Acker, 2004; Jain et al., 1998; Kietzmann et al., 2001). The HIF-1 beta subunit (HIF-1b) is the constitutively expressed heterodimerization partner, which is waiting for the activated HIF-1a to enter the nucleus, where they form the active transcription factor (HIF-1) with additional transcriptional regulators (Prabhakar and Semenza, 2012; Wang et al., 1995). In most cells, the protein ARNT (aryl hydrocarbon receptor nuclear translocator) serves as the HIF-1 beta subunit (Jiang et al., 1996). In addition, a paralogue known as ARNT2 is expressed in the brain (Drutel et al., 1996; Hirose et al., 1996) with a regional pattern (Petersen et al., 2000), which is suggestive of the possibility that many neurons may have

J.A.D. Dela Cruz et al. / Journal of Chemical Neuroanatomy 61–62 (2014) 64–71

two options for the HIF-1 beta subunit. Although ARNT and ARNT2 are encoded by different genes, their high sequence similarity (Drutel et al., 1996; Hirose et al., 1996) has made their study challenging. Since the constitutive expression of different heterodimerization partners is used to diversify cellular responses, a regional difference in HIF-1 response of neurons is of considerable interest. Gene deletion studies in mice and functional studies in vitro have documented only partial overlap between ARNT and ARNT2 in hypoxia regulation (Keith et al., 2001). A recent report of rare ARNT2 mutations in a complex syndrome has proven that ARNT2 has specialized functions in the human brain (Webb et al., 2013), while a genomic study of high altitude populations reported a significant polymorphism of ARNT2, thus suggesting an important physiological role in the hypoxia response in humans (Scheinfeldt et al., 2012). Given that a wealth of findings in the literature has indicated that the dopaminergic neurotransmission system is centrally involved in psychosis (Howes et al., 2013; van Os et al., 2010) and other psychiatric disorders, it is timely to explore the expression of hypoxia-regulated genes such as the HIF-1 in dopaminergic neurons. In this respect, earlier studies have already indicated that hypoxia increases the expression of tyrosine hydroxylase (TH) in peripheral dopaminergic neurons involving HIF-1 (Norris and Millhorn, 1995). Furthermore, evident abnormalities in the levels of the HIF-regulator EGLN1 (PHD2) (Grunblatt, 2004) as well as of glycolytic enzymes under the control of HIF-1 in the substantia nigra (Elstner et al., 2011) have been observed in Parkinson’s disease (i.e. characterized by robust loss of dopaminergic neurons in the midbrain). The expression of ARNT and ARNT2 in dopaminergic neurons has however not been investigated thus far. Here, we studied the constitutively expressed proteins, ARNT and ARNT2, in the substantia nigra pars compacta (SNc), substantia nigra pars reticulata (SNr) and ventral tegmental area (VTA) of mouse using TH as reference marker in immunohistochemistry. Antibodies to ARNT2 were selected by matching of the nuclear labeling pattern obtained by immunohistochemistry with the mRNA in situ hybridization pattern for mouse Arnt2. We also studied proteininteraction networks for ARNT and ARNT2 to determine shared partner proteins.

65

(1:100; Me´dimabs, Montreal, Canada; MM-0063), monoclonal mouse anti-glial fibrillary acidic protein (GFAP) antibody (1:1600; Sigma, St. Louis, MO; G3893) or monoclonal mouse anti-Neuronal Nuclei (NeuN) antibody (1:100; EMD Millipore; Billerica, MA; MAB377) overnight in 0.3% NDS in TBS-T. Slices were incubated with donkey anti-mouse IgG Alexa Fluor 594 (1:100; Invitrogen, Grand Island, NY, USA; a21203) in 0.3% NDS in TBS-T for 2 h. Next, the slices were placed in the second antibody, rabbit anti-Arylhydrocarbon Receptor Nuclear Translocator (ARNT) (1:600; ‘‘HIF1b’’: Novus Biologicals, Cambridge, UK; NB100-110) or rabbit antiAryl Hydrocarbon Receptor Nuclear Translocator 2 (ARNT2) (1:200; Santa Cruz Biotechnology, Inc, Santa Cruz, CA, USA; sc-5581) in 0.3% NDS in TBS-T for overnight incubation. The next day, the secondary antibody, donkey anti-rabbit IgG Alexa Fluor 488 (1:100; Invitrogen, Grand Island, NY; a21206) in 0.3% NDS in TBS-T was applied for 2 h. Lastly, staining with Hoechst dye (1:500; Sigma–Aldrich, Zwijndrecht, The Netherlands) in TBS was carried out for 30 min. Slices were mounted on gelatin coated microscope slides, and coverslipped with glycerol. To control for specificity, a negative control was conducted using a 0.3% NDS in TBST instead of an antibody and for the positive control, the tissue was blocked with 3% NDS. Concurrent single-labeled tissue was also conducted to control for doublelabeling. Immunohistochemistry using peroxidase labeling Endogenous peroxidase activity were quenched by using a 3% hydrogen peroxide solution (Sigma Chemical, St. Louis, MO, USA) for 30 min, slices were blocked for 30 min with 3% NDS in TBS-T, followed by a primary antibody of either Aryl hydrocarbon Receptor Nuclear Translocator (ARNT) (1:600; ‘‘HIF1b’’: Novus Biologicals, Cambridge, UK; NB100-110) or rabbit anti-Aryl Hydrocarbon Receptor Nuclear Translocator 2 (ARNT2) (1:200; Santa Cruz Biotechnology, Inc, Santa Cruz, CA, USA; sc-5581) overnight in 0.3% NDS in TBS-T. Next, slices were incubated in the secondary antibody, donkey anti-rabbit biotin IgG (1:800, Jackson Immunoresearch Laboratories, West Grove, PA; 711-065-152) in 0.3% NDS in TBS-T for 2 h and then placed in avidin–biotin complex for 2 h (1:800, Vector laboratories, UK). To visualize the horseradish peroxide reaction product, the sections were incubated with a 3,30 diaminobenzidine tetrahydrochloride (DAB) (1:1 DAB:Tris?HCl, 0.01% H2O2) (Sigma, Uithoorn, The Netherlands) with nickel chloride enhancement for 10 min and coverslipped using Pertex (HistolabProducts AB, Go¨teborg, Sweden). To control for specificity, a negative control was conducted using a 0.3% NDS in TBST instead of an antibody and for the positive control, the tissue was blocked with 3% NDS. Image analyses For analysis of fluorescence double labeling, a spinning disk confocal microscope (Olympus DSU BX51WI, Pennsylvania, USA) was used. The virtual slice and z-stacks protocols were carried out with a stereological computer microscopy system (Microbrightfield BiosciECe1, Williston, VT). Virtual slices were later stitched on Image J (Preibisch et al., 2009). An Olympus BX-50 microscope (Olympus Nederland BV, Zoeterwoude, Netherlands) was used for capturing images of peroxidase stained sections.

Materials and methods Approval

Stereology

All parts of the experiment involving animals were performed with approval of the Animal Experiments and Ethics Committee of Maastricht University.

Stereological investigation was carried out to evaluate the density of THimmunoreactive neurons containing ARNT-labeled nuclei or ARNT2-labeled nuclei (divided into weakly and strongly labeled) in representative sections of the VTA and SNc. This was done using a stereological computerized microscopy system (Stereo Investigator, Microbrightfield Bioscience, Williston, VT, USA) on a spinning disk confocal microscope (Olympus DSU BX51WI, Pennsylvania, USA). A representative section per mouse midbrain (N = 12; bregma level 3.07 mm; Franklin and Paxinos, 2012) containing the VTA/SNc was delineated and divided in 10,000 mm2 counting frames. Cell counting was performed in all counting frames using the optical fractionator probe. Total cell numbers were estimated by established methods (Schmitz and Hof, 2004; West et al., 1991).

Tissue collection N = 12 wild type male C57Bl6J mice (Central animal facility, Maastricht University, Netherlands) were given an overdose of pentobarbital and underwent perfusion fixation, first with Tyrode’s solution and subsequently with Somogyi solution (4% paraformaldehyde, 15% of a saturated solution of picric acid, 0.05% glutaraldehyde in 0.1 M phosphate buffered saline, PBS). Brains were fixed in fresh fixative (4% paraformaldehyde, 15% picric acid, 0.1 M PBS) at 4 8C for 2 h and then stored in 1% NaN3 at 4 8C. Brains were embedded in 10% gelatin (Sigma–Aldrich, Zwijndrecht, The Netherlands), and cut into 30 mm slices in the frontal plane using a vibratome (Leica1, Wetzlar, Germany). Slices were stored in 1 PBS + 1% NaN3 in a cold room.

Statistical analysis

Because of the sequence similarity of ARNT and ARNT2, a careful selection of antibodies from various commercial sources was carried out. The nuclear labeling pattern for ARNT2 obtained by immunohistochemistry was matched with the mRNA distribution for mouse Arnt2 in coronal sections of the Allen Brain Atlas – http://www.brain-map.org.

All data are presented as means and standard error of means (S.E.M.) Comparisons between groups were performed with an analysis of variance (ANOVA), focusing on the main and interactive effects of the colocalization of TH/ ARNT, TH/ARNT2 (weak) and TH/ARNT2 (strong). Statistical significance is established at p < 0.05. When statistical differences are found, replicate means were compared with a Bonferroni post hoc test for pair-wise comparisons. All statistical calculations were performed using the Statistical Package for Social Sciences (version18 for Apple, SPSS, Chicago, IL, USA).

Immunohistochemistry using fluorescence labeling

Bioinformatics

For double fluorescent visualization of TH and ARNT or ARNT2, a three-day fluorescence protocol was conducted. The slices were blocked in 3% normal donkey serum (NDS) in Tris-buffered saline with 0.3% Triton X-100 (TBS-T) for 30 min, and incubated in the first antibody, monoclonal mouse anti-tyrosine hydroxylase (TH)

Protein–protein interaction (PPI) data were retrieved from GeneMANIA (www.genemania.org), using default settings for all possible interactions, 50 interactions per protein, and pooled data from human, mouse and rat. PPI-datasets for ARNT and ARNT2 were overlapped to determine shared interacting proteins

Selection of ARNT2 antibodies

66

J.A.D. Dela Cruz et al. / Journal of Chemical Neuroanatomy 61–62 (2014) 64–71

with a specific focus on transcription factors. The STRING database (retrieved through IntAct, EMBL-EBI, www.ebl.ac.uk/intact/main.xhtml) served as a second source for PPIs. Literature searches were used to supplement this analysis.

Results ARNT General distribution ARNT antibodies showed strong and widespread neuronal labeling without evident regional differences on inspection. The hippocampus and neocortex revealed dense neuronal labeling (Fig. 1A and B). Nuclear labeling was granular, with sparing of the nucleolus, and cytoplasmic staining was also present. The generalized labeling pattern with antibodies to ARNT is in line with the description of global expression of ARNT mRNA (Aitola and Pelto-Huikko, 2003; Kainu et al., 1995; Petersen et al., 2000). ARNT antibodies showed widespread labeling in the mouse midbrain, including substantia nigra (Fig. 1C). ARNT/TH double-staining We focused on the dopaminergic neurons of the midbrain, i.e. SNc and VTA (Fig. 2A). Labeling for ARNT was seen in the nucleus of virtually all TH-immunoreactive neurons in the SNc and VTA (Fig. 2E–G). Co-localization of ARNT and TH staining was limited to the cytoplasm (Fig. 2B–D). ARNT2

most of the thalamus (Fig. 3B). In the midbrain, a cluster of strong ARNT2 labeling was seen in the SNr (Fig. 3C; see also Fig. 5). The regional distribution of strong nuclear labeling for ARNT2 showed good correspondence with the distribution maps for Arnt2 mRNA (Allen Brain Atlas; Lein et al., 2007), particularly in the hippocampus. Thus, strong labeling for ARNT2 can be interpreted with confidence. ARNT2/TH double-staining TH immunohistochemistry was used to delineate the dopaminergic neurons in SNc and VTA (Fig. 5A). Nuclei strongly labeled for ARNT2 were clustered in SNr outside the TH-immunoreactive zone (Fig. 5A). In addition, some nuclei strongly labeled for ARNT2 were scattered throughout the SNc and few were observed in the VTA (Fig. 5A). At higher magnification, weak immunolabeling for ARNT2 was observed in TH-immunoreactive neurons of SNc (Fig. 5F) and VTA (Fig. 5G). In the SNr, nuclei with strong ARNT2 single labeling were found (Fig. 5E). In the SNr, SNc and VTA, nuclei strongly labeled for ARNT2 were not surrounded by TH-immunoreactive cytoplasm (Fig. 5B–E). ARNT2/NeuN and ARNT2/GFAP co-localization Double-labeling of nuclei strongly labeled for ARNT2 in the SNr with the neuronal marker NeuN indicated that these nuclei are in neurons (Fig. 6A–C). Nuclear ARNT2-labeling was not found to co-localize with GFAP as an astrocytic marker (Fig. 6D–F). Stereology

General distribution ARNT2 immunohistochemistry showed two patterns of labeling. Strong nuclear immunolabeling for ARNT2 was seen in the cortex (Figs. 3A–C and 4A), CA1 (Figs. 3B and 4B), in cells scattered in the dendritic layers of the hippocampus (Fig. 3B), in the hilar region (Fig. 4C), the reticular thalamic nucleus (Figs. 3B and 4D) and the globus pallidus. Immunolabeling for ARNT2 was very low in the striatum (Fig. 3A), CA3 (Fig. 3B), the dentate gyrus and in

We focused the stereological analyses on the expression of ARNT or ARNT2 in dopaminergic midbrain neurons that were labeled by TH-antibodies in their cytoplasm. A one-way ANOVA indicated that in the VTA, the colocalization of TH-immunoreactive neurons with ARNT-positive nuclei, the colocalization of THimmunoreactive neurons with nuclei weakly labeled for ARNT2, and the colocalization of TH-immunoreactive neurons with

Fig. 1. (A) An overview of a coronal section at the level of the midbrain (bregma level 3.07 mm), immunolabeled for ARNT. (B) A detailed view of the dentate gyrus of the hippocampus. (C) A detailed view of the substantia nigra. Scale bar = 500 mm in A and 100 mm in B–C.

J.A.D. Dela Cruz et al. / Journal of Chemical Neuroanatomy 61–62 (2014) 64–71

67

Fig. 2. (A) An overview of the midbrain at bregma level 3.15 mm immunolabeled for ARNT in green and for TH in red in the substantia nigra pars compacta (SNc) and the ventral tegmental area (VTA). The medial lemniscus (ML) separates the two regions. (B–D) Images show three-dimensional reconstructions of representative image stacks. The crossing lines indicate the position within the X–Y view at which the Y–Z and X–Z views were generated. (E–F) Detailed views of the (E) SNr, (F) SNc, and (G) VTA. Thick white arrows point to representative colocalization between ARNT (green) and TH (red). The thin yellow arrow is an example of a single-labeled ARNT-immunoreactive cell. Scale bar = 100 mm in A and 20 mm in B–G. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

nuclei strongly labeled for ARNT2 was significantly different (F(2,15) = 126.774, p < 0.05). Pair-wise post hoc comparisons using Bonferroni correction showed that the mean proportion of TH-immunoreactive neurons with ARNT-immunoreactive nuclei

(M = 0.8282; SEM = 0.03593) was significantly larger than the mean proportion of TH-immunoreactive neurons with nuclei weakly labeled for ARNT2 (M = 0.5877; SEM = 0.05447) or nuclei strongly labeled for ARNT2 (M = 0).

Fig. 3. Low powered views of sections with peroxidase labeling for ARNT2 at bregma levels of (A) 0.47 mm, (B) 1.79 mm, (C) 3.15 mm. Scale bar = 500 mm. Boxes labeled with a–d indicate areas shown at higher magnification in Fig. 4.

68

J.A.D. Dela Cruz et al. / Journal of Chemical Neuroanatomy 61–62 (2014) 64–71

Fig. 4. Detailed views of sections with peroxidase labeling for ARNT2 in areas indicated from Fig. 3. (A) Cortex at bregma level 0.47 mm, (B) CA1 of the hippocampus at bregma level 1.79 mm, (C) the hilus of the dentate gyrus at bregma level 1.79 mm, and (D) reticular thalamic nucleus at bregma level 1.79 mm. Scale bar = 100 mm.

In the SNc, a one-way ANOVA indicated that the colocalization of the TH-immunoreactive neurons with nuclei labeled for ARNT, the colocalization of TH-immunoreactive neurons with nuclei weakly labeled for ARNT2, and the colocalization of TH-immunoreactive neurons with nuclei strongly labeled for ARNT2 showed was significantly different (F(2,15) = 62.955, p < 0.05). Pair-wise post hoc comparisons using Bonferroni correction showed that the mean proportion of TH-immunoreactive neurons coinciding with ARNT-immunoreactive nuclei (M = 0.8625; SEM = 0.02462) was significantly larger than the mean proportion of TH-immunoreactive neurons coinciding with nuclei weakly labeled for ARNT2 (M = 0.5022; SEM = 0.09170) and the mean proportion of THimmunoreactive cells coinciding with nuclei strongly labeled for ARNT2 (M = 0). In summary, more than 80% of the TH-immunoreactive neurons contained nuclei with ARNT-immunoreactivity, while about 55% of TH-immunoreactive neurons had nuclei weakly labeled for ARNT2. Nuclei strongly labeled for ARNT2 were not observed in THimmunoreactive neurons (0%). Bioinformatics PPIs obtained from GeneMANIA and STRING revealed shared protein partners for ARNT and ARNT2, i.e. AHR, EPAS1 (HIF2A), HIF1A, NPAS4, POU3F2, SIM1 and SIM2. The STRING database listed another 164 putative partner proteins shared by ARNT and ARNT2. Two interactions for transcription factors relevant for dopaminergic neurons were found (STRING), i.e. FOXA2 with ARNT, and PITX2 with both ARNT and ARNT2. Another major transcription factor in dopaminergic neurons, NR4A2 (NURR1), has been listed for HIF-1 regulation in a promoter analysis (Scho¨del et al., 2011). BHLHE40 (DEC1) and NPAS1 were specific

for interaction with ARNT; interestingly, NPAS1 has been shown to suppress TH (Teh et al., 2007).

Discussion In order to address the expression of genes regulating the response to hypoxia in dopaminergic neurons in the SNc and VTA, we performed neuroanatomical studies on regional and cell-typespecific expression of ARNT and ARNT2 in the mouse midbrain. Most TH-immunoreactive dopaminergic neurons in SNc and VTA showed labeling for ARNT in their nucleus. ARNT immunoreactivity was similarly seen in SNr and in other neurons in the forebrain, which is in line with the generalized expression of ARNT mRNA as published previously (Petersen et al., 2000). We found regional patterns of nuclei with strong labeling for ARNT2 in the forebrain, which is congruent with earlier observations on the Arnt2 mRNA distribution (Petersen et al., 2000). Our double-labeling studies indicated that nuclei strongly labeled for ARNT2 belong to neurons. Specifically, nuclei strongly labeled for ARNT2 were found clustered in the SNr, but rarely observed in SNc or VTA. Thus, we observed distinct co-localization patterns of TH with ARNT and ARNT2 in the mouse midbrain that were confirmed by stereological analysis. Our data suggest that most dopaminergic neurons in SNc may be limited to ARNT as partner for HIF-1a during hypoxia responses. This observation may be related to the hypothesis that their large axonal arbor puts SNc neurons on a tight energy budget (Bolam and Pissadaki, 2012). Previous studies have focused almost entirely on the oxygenregulated protein HIF-1 alpha that is ubiquitously expressed by neurons (Kietzmann et al., 2001; Acker and Acker, 2004). Our observations of region-, and cell-type-specific co-localization

J.A.D. Dela Cruz et al. / Journal of Chemical Neuroanatomy 61–62 (2014) 64–71

69

Fig. 5. (A) An overview of the midbrain at bregma level 3.07 mm showing the ARNT2 signals in green and TH signals in red in the SN and the VTA. The medial lemniscus (ML) separates the two regions. (B–D) Images show a three-dimensional reconstruction of a representative image stack. The crossing lines indicate the position within the X–Y view at which the Y–Z and X–Z views were generated. (E–F) High powered views of the (E) SNr, (F) SNc and (G) VTA. Thick white arrows represent the double-labeling of TH in the cytoplasm and nuclei with weak labeling for ARNT2. Thin yellow arrows represent examples of nuclei strongly labeled for ARNT2. The asterisk (*) represents a THimmunoreactive neuron devoid of ARNT2 immunoreactivity in the nucleus. Scale bar = 100 mm in A and 20 mm in B–G. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

patterns on the constitutively expressed HIF-1 beta proteins may be of functional importance, because ARNT and ARNT2 are known to have only partially overlapping functions in hypoxia signaling (Keith et al., 2001). Therefore, distinct brain regions and cell types may respond differently on hypoxic challenges, based on differences in ARNT and ARNT2 expression (Maltepe et al., 2000; Stolze et al., 2002). Distinctly different patterns of expression for ARNT and ARNT2 are also found during brain development (Aitola and Pelto-Huikko, 2003). Gene deletion of Arnt2 in the mouse has lead to hypothalamic abnormalities and premature postnatal death. These were specifically related to SIM1, and not HIF-1 functions (Michaud et al., 2000). Because ARNT/HIF-1 alpha complexes are still formed in Arnt2 / neurons, these may be sufficient to maintain normal steady-state levels of target gene expression in vivo. Accordingly, the vascularization in the Arnt2 / brains was normal during development, whereas the deletion of Arnt had dramatic consequences for vascular development (Keith et al., 2001). These

authors suggested that ARNT2/HIF-1 alpha can regulate HIF-1 target genes, but ARNT/HIF-1 alpha complexes may be sufficient to regulate normal hypoxic responses during development in vivo (Keith et al., 2001). Other studies postulated functional differences in the role of ARNT and ARNT2 in neurons (Drutel et al., 1999, 2000). Observations in cell culture and in a brain ischemia model suggest that ARNT2 plays a role in neuronal survival (Drutel et al., 1999, 2000). In addition to these studies, overlapping roles were found for ARNT and ARNT2 in the hypoxia response of transient and stable expression systems (Sekine et al., 2006). Recently, Hao et al. (2013) showed high Arnt/low Arnt2 expression in undifferentiated cells and a strong increase in Arnt2 upon neuronal differentiation, while the switch for high Arnt2 was regulated by epigenetic modifications in neuronal differentiation (Hao et al., 2013). It has been proposed that expression of ARNT or ARNT2 allows for the formation of differential transcriptional complexes in neurons (Michaud et al., 2000; Hao et al., 2013). Differences between expression of either ARNT alone or the combination of

70

J.A.D. Dela Cruz et al. / Journal of Chemical Neuroanatomy 61–62 (2014) 64–71

Fig. 6. (A) Immunohistochemical staining of neurons using Neuronal Nuclei (NeuN) antibody (red), (B) immunohistochemical staining with ARNT2 antibody (green) and (C) NeuN/ARNT2 labeling. (D) Immunohistochemical staining of astrocytes using glial fibrillary acidic protein (GFAP) antibody (red), (E) immunohistochemical staining with ARNT2 antibody (green), and (F) GFAP/ARNT2 labeling. (A–C) and (D–F) Images show three-dimensional reconstructions of representative image stacks. The crossing lines indicate the position within the X–Y view at which the Y–Z and X–Z views were generated. Scale bar = 20 mm. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

ARNT and ARNT2 were estimated here from protein interaction data. ARNT and ARNT2 can form heterodimers with other basic helix-loop-helix (bHLH) transcription factors (Aitola and PeltoHuikko, 2003), i.e. AHR, EPAS1 (HIF2A), NPAS4, SIM1, and SIM2. AHR is a receptor for environmental toxins of the dioxin group (McIntosh et al., 2010), and AHR has also been characterized as a receptor for endogenous ligands, including kynurenine (Opitz et al., 2011). ARNT2 appears to be less important than ARNT as a partner for AHR (Dougherty and Pollenz, 2008; Sekine et al., 2006), which could be due to the presence of inhibitory proteins (Hao et al., 2013). AHR and ARNT are upregulated by dioxin in the brain including the substantia nigra (Huang et al., 2000). Interestingly, interactions between dioxin and tyrosine hydroxylase have been demonstrated that involve AHR (Akahoshi et al., 2009). Furthermore, interactions between HIF-1 and AHR-signaling via ARNT as shared heterodimerization partner have been described (Fleming et al., 2009). It is possible that the expression of ARNT2 in adult dopaminergic midbrain neurons is limited by design in order to dampen hypoxia responses. TH has been described as hypoxia regulated by involving HIF-1 in peripheral dopaminergic neurons (Mannello et al., 2011; Norris and Millhorn, 1995; Panchision, 2009). It can thus be speculated that dopaminergic neurons may want to avoid fluctuations of TH expression due to local hypoxic events in midbrain because it would impact the dopamine levels in the terminals in the remote striatum. In fact, NPAS1 has been shown to suppress TH (Teh et al., 2007) and, according to our analysis of protein interactions, NPAS1 binds with ARNT but not ARNT2. The present findings are of relevance for genetic studies in psychiatric and neurological disorders. A recent report of ARNT2 mutations leading to loss function indicates important roles in postnatal brain development (Webb et al., 2013). ARNT2 has been also reported to interact with key proteins in neurodevelopmental disorders, i.e. necdin in Prader–Willi syndrome (Friedman and Fan, 2007) and FMRP in Fragile X-Syndrome (Darnell et al., 2011). Given the low ARNT2 expression, genetic or epigenetic variation in ARNT could be an important factor in the hypoxia response of dopaminergic midbrain neurons. Two studies in schizophrenia analyzed methylation patterns and found alterations of ARNT and EPAS1 (HIF2A) (Aberg et al., 2014; Wockner et al., 2014). Limited options for HIF-1 responses in SNc neurons have also implications for degeneration in Parkinson’s disease. To conclude, our observations of distinct co-localization patterns of TH with ARNT and ARNT2 in the mouse midbrain may serve as a starting point for further studies on the effects of

hypoxia on dopaminergic systems and the biological underpinnings of the impact of hypoxia in the etiology of a range of psychiatric and neurological disorders. Acknowledgements Many thanks to Hellen Steinbusch, Roy Lardenoije and Lieke van Gastel for their technical support on this project. BPFR has received funding from the European Community’s Seventh Framework Programme under Grant Agreement No. HEALTH-F22009-241909 (Project EU-GEI) and from the Netherlands Organisation for Scientific Research (NWO; VENI Award No. 916.11.086). The micrographs in this paper were taken with a confocal spinning disk microscope financed by the Netherlands Organisation for Scientific Research (NWO), grant number 911-06-003.

References Aberg, K.A., McClay, J.L., Nerella, S., Clark, S., Kumar, G., Chen, W., Khachane, A.N., Xie, L., Hudson, A., Gao, G., Harada, A., Hultman, C.M., Sullivan, P.F., Magnusson, P.K., van den Oord, E.J., 2014. Methylome-wide association study of schizophrenia: identifying blood biomarker signatures of environmental insults. JAMA Psychiatry 71 (3) 255–264. Acker, T., Acker, H., 2004. Cellular oxygen sensing need in CNS function: physiological and pathological implications. J. Exp. Biol. 207, 3171–3188. Aitola, M.H., Pelto-Huikko, M.T., 2003. Expression of Arnt and Arnt2 mRNA in developing murine tissues. J. Histochem. Cytochem. 51, 41–54. Akahoshi, E., Yoshimura, S., Uruno, S., Ishihara-Sugano, M., 2009. Effect of dioxins on regulation of tyrosine hydroxylase gene expression by aryl hydrocarbon receptor: a neurotoxicology study. Environ. Health 8, 24. Bolam, J.P., Pissadaki, E.K., 2012. Living on the edge with too many mouths to feed: why dopamine neurons die. Mov. Disord. 27, 1478–1483. Darnell, J.C., Van Driesche, S.J., Zhang, C., Hung, K.Y., Mele, A., Fraser, C.E., Stone, E.F., Chen, C., Fak, J.J., Chi, S.W., Licatalosi, D.D., Richter, J.D., Darnell, R.B., 2011. FMRP stalls ribosomal translocation on mRNAs linked to synaptic function and autism. Cell 146, 247–261. Dougherty, E.J., Pollenz, R.S., 2008. Analysis of Ah receptor-ARNT and Ah receptorARNT2 complexes in vitro and in cell culture. Toxicol. Sci. 103, 191–206. Drutel, G., Heron, A., Kathmann, M., Gros, C., Mace, S., Plotkine, M., Schwartz, J.C., Arrang, J.M., 1999. ARNT2, a transcription factor for brain neuron survival? Eur. J. Neurosci. 11, 1545–1553. Drutel, G., Kathmann, M., Heron, A., Gros, C., Mace, S., Schwartz, J.C., Arrang, J.M., 2000. Two splice variants of the hypoxia-inducible factor HIF-1alpha as potential dimerization partners of ARNT2 in neurons. Eur. J. Neurosci. 12, 3701–3708. Drutel, G., Kathmann, M., Heron, A., Schwartz, J.C., Arrang, J.M., 1996. Cloning and selective expression in brain and kidney of ARNT2 homologous to the Ah receptor nuclear translocator (ARNT). Biochem. Biophys. Res. Commun. 225, 333–339. Elstner, M., Morris, C.M., Heim, K., Bender, A., Mehta, D., Jaros, E., Klopstock, T., Meitinger, T., Turnbull, D.M., Prokisch, H., 2011. Expression analysis of dopaminergic neurons in Parkinson’s disease and aging links transcriptional dysregulation of energy metabolism to cell death. Acta Neuropathol. 122, 75–86.

J.A.D. Dela Cruz et al. / Journal of Chemical Neuroanatomy 61–62 (2014) 64–71 Fleming, C.R., Billiard, S.M., Di Giulio, R.T., 2009. Hypoxia inhibits induction of aryl hydrocarbon receptor activity in topminnow hepatocarcinoma cells in an ARNT-dependent manner. Comp. Biochem. Physiol. C: Toxicol. Pharmacol. 150, 383–389. Franklin, K.B.J., Paxinos, G., 2012. The Mouse Brain in Stereotaxic Coordinates Fourth Edition. Academic Press, New York. Friedman, E.R., Fan, C.M., 2007. Separate necdin domains bind ARNT2 and HIF1alpha and repress transcription. Biochem. Biophys. Res. Commun. 363, 113–118. Gilman, S.R., Chang, J., Xu, B., Bawa, T.S., Gogos, J.A., Karayiorgou, M., Vitkup, D., 2012. Diverse types of genetic variation converge on functional gene networks involved in schizophrenia. Nat. Neurosci. 15, 1723–1728. Grunblatt, E., 2004. The benefits of microarrays as tools for studying neuropsychiatric disorders. Drugs Today (Barc.) 40, 147–156. Gulsuner, S., Walsh, T., Watts, A.C., Lee, M.K., Thornton, A.M., Casadei, S., Rippey, C., Shahin, H., Consortium on the Genetics of Schizophrenia (COGS), PAARTNERS Study Group, Nimgaonkar, V.L., Go, R.C., Savage, R.M., Swerdlow, N.R., Gur, R.E., Braff, D.L., King, M.C., McClellan, J.M., 2013. Spatial and temporal mapping of de novo mutations in schizophrenia to a fetal prefrontal cortical network. Cell 154 (3) 518–529. Hao, N., Bhakti, V.L., Peet, D.J., Whitelaw, M.L., 2013. Reciprocal regulation of the basic helix-loop-helix/Per-Arnt-Sim partner proteins Arnt and Arnt2, during neuronal differentiation. Nucleic Acids Res. 41, 5626–5638. Hirose, K., Morita, M., Ema, M., Mimura, J., Hamada, H., Fujii, H., Saijo, Y., Gotoh, O., Sogawa, K., Fujii-Kuriyama, Y., 1996. cDNA cloning and tissue-specific expression of a novel basic helix-loop-helix/PAS factor (Arnt2) with close sequence similarity to the aryl hydrocarbon receptor nuclear translocator (Arnt). Mol. Cell. Biol. 16, 1706–1713. Howes, O.D., Williams, M., Ibrahim, K., Leung, G., Egerton, A., McGuire, P.K., Turkheimer, F., 2013. Midbrain dopamine function in schizophrenia and depression: a post-mortem and positron emission tomographic imaging study. Brain 136, 3242–3251. Huang, P., Rannug, A., Ahlbom, E., Hakansson, H., Ceccatelli, S., 2000. Effect of 2,3,7,8-tetrachlorodibenzo-p-dioxin on the expression of cytochrome P450 1A1, the aryl hydrocarbon receptor, and the aryl hydrocarbon receptor nuclear translocator in rat brain and pituitary. Toxicol. Appl. Pharmacol. 169, 159–167. Jain, S., Maltepe, E., Lu, M.M., Simon, C., Bradfield, C.A., 1998. Expression of ARNT,ARNT2, HIF1 alpha HIF2 alpha and Ah receptor mRNAs in the developing mouse. Mech. Dev. 73, 117–123. Jiang, B.H., Rue, E., Wang, G.L., Roe, R., Semenza, G.L., 1996. Dimerization, DNA binding, and transactivation properties of hypoxia-inducible factor 1. J. Biol. Chem. 271, 17771–17778. Kainu, T., Gustafsson, J.A., Pelto-Huikko, M., 1995. The dioxin receptor and its nuclear translocator (Arnt) in the rat brain. Neuroreport 6, 2557–2560. Keith, B., Adelman, D.M., Simon, M.C., 2001. Targeted mutation of the murine arylhydrocarbon receptor nuclear translocator 2 (Arnt2) gene reveals partial redundancy with Arnt. Proc. Natl. Acad. Sci. U. S. A. 98, 6692–6697. Kietzmann, T., Knabe, W., Schmidt-Kastner, R., 2001. Hypoxia and hypoxia-inducible factor modulated gene expression in brain: involvement in neuroprotection and cell death. Eur. Arch. Psychiatry Clin. Neurosci. 251, 170–178. Lein, E.S., Hawrylycz, M.J., Ao, N., Ayres, M., Bensinger, A., Bernard, A., Boe, A.F., Boguski, M.S., Brockway, K.S., Byrnes, E.J., Chen, L., Chen, L., Chen, T.M., Chin, M.C., Chong, J., Crook, B.E., Czaplinska, A., Dang, C.N., Datta, S., Dee, N.R., Desaki, A.L., Desta, T., Diep, E., Dolbeare, T.A., Donelan, M.J., Dong, H.W., Dougherty, J.G., Duncan, B.J., Ebbert, A.J., Eichele, G., Estin, L.K., Faber, C., Facer, B.A., Fields, R., Fischer, S.R., Fliss, T.P., Frensley, C., Gates, S.N., Glattfelder, K.J., Halverson, K.R., Hart, M.R., Hohmann, J.G., Howell, M.P., Jeung, D.P., Johnson, R.A., Karr, P.T., Kawal, R., Kidney, J.M., Knapik, R.H., Kuan, C.L., Lake, J.H., Laramee, A.R., Larsen, K.D., Lau, C., Lemon, T.A., Liang, A.J., Liu, Y., Luong, L.T., Michaels, J., Morgan, J.J., Morgan, R.J., Mortrud, M.T., Mosqueda, N.F., Ng, L.L., Ng, R., Orta, G.J., Overly, C.C., Pak, T.H., Parry, S.E., Pathak, S.D., Pearson, O.C., Puchalski, R.B., Riley, Z.L., Rockett, H.R., Rowland, S.A., Royall, J.J., Ruiz, M.J., Sarno, N.R., Schaffnit, K., Shapovalova, N.V., Sivisay, T., Slaughterbeck, C.R., Smith, S.C., Smith, K.A., Smith, B.I., Sodt, A.J., Stewart, N.N., Stumpf, K.R., Sunkin, S.M., Sutram, M., Tam, A., Teemer, C.D., Thaller, C., Thompson, C.L., Varnam, L.R., Visel, A., Whitlock, R.M., Wohnoutka, P.E., Wolkey, C.K., Wong, V.Y., Wood, M., Yaylaoglu, M.B., Young, R.C., Youngstrom, B.L., Yuan, X.F., Zhang, B., Zwingman, T.A., Jones, A.R., 2007. Genome-wide atlas of gene expression in the adult mouse brain. Nature 445 (7124) 168–176. Maltepe, E., Keith, B., Arsham, A.M., Brorson, J.R., Simon, M.C., 2000. The role of ARNT2 in tumor angiogenesis and the neural response to hypoxia. Biochem. Biophys. Res. Commun. 273, 231–238. Mannello, F., Medda, V., Tonti, G.A., 2011. Hypoxia and neural stem cells: from invertebrates to brain cancer stem cells. Int. J. Dev. Biol. 55, 569–581. McIntosh, B.E., Hogenesch, J.B., Bradfield, C.A., 2010. Mammalian Per-Arnt-Sim proteins in environmental adaptation. Annu. Rev. Physiol. 72, 625–645. Michaud, J.L., DeRossi, C., May, N.R., Holdener, B.C., Fan, C.M., 2000. ARNT2 acts as the dimerization partner of SIM1 for the development of the hypothalamus. Mech. Dev. 90, 253–261. Neale, B.M., Kou, Y., Liu, L., Ma’ayan, A., Samocha, K.E., Sabo, A., Lin, C.F., Stevens, C., Wang, L.S., Makarov, V., Polak, P., Yoon, S., Maguire, J., Crawford, E.L., Campbell, N.G., Geller, E.T., Valladares, O., Schafer, C., Liu, H., Zhao, T., Cai, G., Lihm, J., Dannenfelser, R., Jabado, O., Peralta, Z., Nagaswamy, U., Muzny, D., Reid, J.G., Newsham, I., Wu, Y., Lewis, L., Han, Y., Voight, B.F., Lim, E., Rossin, E., Kirby, A.,

71

Flannick, J., Fromer, M., Shakir, K., Fennell, T., Garimella, K., Banks, E., Poplin, R., Gabriel, S., DePristo, M., Wimbish, J.R., Boone, B.E., Levy, S.E., Betancur, C., Sunyaev, S., Boerwinkle, E., Buxbaum, J.D., Cook Jr., E.H., Devlin, B., Gibbs, R.A., Roeder, K., Schellenberg, G.D., Sutcliffe, J.S., Daly, M.J., 2012. Patterns and rates of exonic de novo mutations in autism spectrum disorders. Nature 485, 242–245. Norris, M.L., Millhorn, D.E., 1995. Hypoxia-induced protein binding to O2-responsive sequences on the tyrosine hydroxylase gene. J. Biol. Chem. 270, 23774– 23779. O’Roak, B.J., Vives, L., Girirajan, S., Karakoc, E., Krumm, N., Coe, B.P., Levy, R., Ko, A., Lee, C., Smith, J.D., Turner, E.H., Stanaway, I.B., Vernot, B., Malig, M., Baker, C., Reilly, B., Akey, J.M., Borenstein, E., Rieder, M.J., Nickerson, D.A., Bernier, R., Shendure, J., Eichler, E.E., 2012. Sporadic autism exomes reveal a highly interconnected protein network of de novo mutations. Nature 485, 246–250. Opitz, C.A., Litzenburger, U.M., Sahm, F., Ott, M., Tritschler, I., Trump, S., Schumacher, T., Jestaedt, L., Schrenk, D., Weller, M., Jugold, M., Guillemin, G.J., Miller, C.L., Lutz, C., Radlwimmer, B., Lehmann, I., von Deimling, A., Wick, W., Platten, M., 2011. An endogenous tumour-promoting ligand of the human aryl hydrocarbon receptor. Nature 478, 197–203. Panchision, D.M., 2009. The role of oxygen in regulating neural stem cells in development and disease. J. Cell. Physiol. 220, 562–568. Petersen, S.L., Curran, M.A., Marconi, S.A., Carpenter, C.D., Lubbers, L.S., McAbee, M.D., 2000. Distribution of mRNAs encoding the arylhydrocarbon receptor, arylhydrocarbon receptor nuclear translocator, and arylhydrocarbon receptor nuclear translocator-2 in the rat brain and brainstem. J. Comp. Neurol. 427, 428–439. Prabhakar, N.R., Semenza, G.L., 2012. Adaptive and maladaptive cardiorespiratory responses to continuous and intermittent hypoxia mediated by hypoxia-inducible factors 1 and 2. Physiol. Rev. 92, 967–1003. Preibisch, S., Saalfeld, S., Tomancak, P., 2009. Globally optimal stitching of tiled 3D microscopic image acquisitions. Bioinformatics 25, 1463–1465. Richards, A.L., Jones, L., Moskvina, V., Kirov, G., Gejman, P.V., Levinson, D.F., Sanders, A.R., Molecular Genetics of Schizophrenia, C., International Schizophrenia, C., Purcell, S., Visscher, P.M., Craddock, N., Owen, M.J., Holmans, P., O’Donovan, M.C., 2012. Schizophrenia susceptibility alleles are enriched for alleles that affect gene expression in adult human brain. Mol. Psychiatry 17, 193–201. Scheinfeldt, L.B., Soi, S., Thompson, S., Ranciaro, A., Woldemeskel, D., Beggs, W., Lambert, C., Jarvis, J.P., Abate, D., Belay, G., Tishkoff, S.A., 2012. Genetic adaptation to high altitude in the Ethiopian highlands. Genome Biol. 13, R1. Scho¨del, J., Oikonomopoulos, S., Ragoussis, J., Pugh, C.W., Ratcliffe, P.J., Mole, D.R., 2011. High-resolution genome-wide mapping of HIF-binding sites by ChIP-seq. Blood 117 (23) e207–e217. Schmidt-Kastner, R., van Os, J., Esquivel, G., Steinbusch, H.W., Rutten, B.P., 2012. An environmental analysis of genes associated with schizophrenia: hypoxia and vascular factors as interacting elements in the neurodevelopmental model. Mol. Psychiatry 17, 1194–1205. Schmitz, C., Hof, P.R., 2004. Design-based stereology in neuroscience. Neuroscience 130 (4) 813–831. Sekine, H., Mimura, J., Yamamoto, M., Fujii-Kuriyama, Y., 2006. Unique and overlapping transcriptional roles of arylhydrocarbon receptor nuclear translocator (Arnt) and Arnt2 in xenobiotic and hypoxic responses. J. Biol. Chem. 281, 37507–37516. Sharp, F.R., Bernaudin, M., 2004. HIF1 and oxygen sensing in the brain. Nat. Rev. Neurosci. 5, 437–448. Shibata, T., Yamagata, H., Uchida, S., Otsuki, K., Hobara, T., Higuchi, F., Abe, N., Watanabe, Y., 2013. The alteration of hypoxia inducible factor-1 (HIF-1) and its target genes in mood disorder patients. Prog. Neuropsychopharmacol. Biol. Psychiatry 43, 222–229. Stolze, I., Berchner-Pfannschmidt, U., Freitag, P., Wotzlaw, C., Rossler, J., Frede, S., Acker, H., Fandrey, J., 2002. Hypoxia-inducible erythropoietin gene expression in human neuroblastoma cells. Blood 100, 2623–2628. Teh, C.H., Loh, C.C., Lam, K.K., Loo, J.M., Yan, T., Lim, T.M., 2007. Neuronal PAS domain protein 1 regulates tyrosine hydroxylase level in dopaminergic neurons. J .Neurosci. Res. 85, 1762–1773. van Os, J., Kenis, G., Rutten, B.P., 2010. The environment and schizophrenia. Nature 468, 203–212. Wang, G.L., Jiang, B.H., Rue, E.A., Semenza, G.L., 1995. Hypoxia-inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension. Proc. Natl. Acad. Sci. U. S. A. 92, 5510–5514. Webb, E.A., AlMutair, A., Kelberman, D., Bacchelli, C., Chanudet, E., Lescai, F., Andoniadou, C.L., Banyan, A., Alsawaid, A., Alrifai, M.T., Alahmesh, M.A., Balwi, M., Mousavy-Gharavy, S.N., Lukovic, B., Burke, D., McCabe, M.J., Kasia, T., Kleta, R., Stupka, E., Beales, P.L., Thompson, D.A., Chong, W.K., Alkuraya, F.S., MartinezBarbera, J.P., Sowden, J.C., Dattani, M.T., 2013. ARNT2 mutation causes hypopituitarism, post-natal microcephaly, visual and renal anomalies. Brain 136, 3096–3105. West, M.J., Slomianka, L., Gundersen, H.J., 1991. Unbiased stereological estimation of the total number of neurons in thesubdivisions of the rat hippocampus using the optical fractionator. Anat. Rec. 231, 482–497. Wockner, L.F., Noble, E.P., Lawford, B.R., Young, R.M., Morris, C.P., Whitehall, V.L., Voisey, J., 2014. Genome-wide DNA methylation analysis of human brain tissue from schizophrenia patients. Transl. Psychiatry 4, e339.