Author's Accepted Manuscript

Impaired spatial learning and reduced adult hippocampal neurogenesis in histamine H1receptor knockout mice Oliver Ambrée, Jens Buschert, Weiqi Zhang, Volker Arolt, Ekrem Dere, Armin Zlomuzica

www.elsevier.com/locate/euroneuro

PII: DOI: Reference:

S0924-977X(14)00128-X http://dx.doi.org/10.1016/j.euroneuro.2014.04.006 NEUPSY10829

To appear in:

European Neuropsychopharmacology

Received date: 11 February 2014 Revised date: 3 April 2014 Accepted date: 27 April 2014 Cite this article as: Oliver Ambrée, Jens Buschert, Weiqi Zhang, Volker Arolt, Ekrem Dere, Armin Zlomuzica, Impaired spatial learning and reduced adult hippocampal neurogenesis in histamine H1-receptor knockout mice, European Neuropsychopharmacology, http://dx.doi.org/10.1016/j.euroneuro.2014.04.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Impaired spatial learning and reduced adult hippocampal neurogenesis in histamine H1receptor knockout mice Oliver Ambrée1, Jens Buschert1, Weiqi Zhang1, Volker Arolt1, Ekrem Dere2,3,4, Armin Zlomuzica*2,5 1

Department of Psychiatry, University of Münster, Germany

2

Institute of Physiological Psychology, Heinrich-Heine University, Düsseldorf, Germany

3

UMR 7102, Neurobiologie des Processus Adaptatifs, Université Pierre et Marie Curie, Paris 6, France 4

Max Planck Institute of Experimental Medicine, Göttingen, Germany

5

Mental Health Research and Treatment Center, University of Bochum, Germany

*Corresponding author: A. Zlomuzica, Mental Health Research and Treatment Center, University of Bochum, Germany. E-mail: [email protected]

Abstract The histamine H1-receptor (H1R) is expressed in wide parts of the brain including the hippocampus, which is involved in spatial learning and memory. Previous studies in H1R knockout (KO) mice revealed deficits in a variety of learning and memory tasks. It was also proposed that H1R activation is crucial for neuronal differentiation of neural progenitors. Therefore, the aim of this study was to investigate negatively reinforced spatial learning in the water-maze and to assess survival and neuronal differentiation of newborn cells in the adult 1

hippocampus of H1R-KO mice. H1R-KO and wild-type (WT) mice were subjected to the following sequence of tests a) cued version, b) place learning, c) spatial probe, d) long-term retention and e) reversal learning. Furthermore hippocampal neurogenesis in terms of survival and differentiation was assessed in H1R-KO and WT mice. H1R-KO mice showed normal cued learning, but impaired place and reversal learning as well as impaired long-term retention performance. In addition, a marked reduction of newborn neurons in the hippocampus but no changes in differentiation of neural progenitors into neuronal and glial lineage was found in H1R-KO mice. Our data suggest that H1R deficiency in mice is associated with pronounced deficits in hippocampusdependent spatial learning and memory. Furthermore, we herein provide first evidence that H1R deficiency in the mouse leads to a reduced neurogenesis. However, the exact mechanisms for the reduced number of cells in H1R KO mice remain elusive and might be due to a reduced survival of newborn hippocampal neurons and/or a reduction in cell proliferation.

1. Introduction Histamine synthesizing neurons have been exclusively identified in the nucleus tuberomammillaris in the posterior part of the hypothalamus. The nucleus tuberomammillaris receives inputs from limbic areas and diffusely innervates wide parts of the brain (Haas et al., 2008). Histamine synthesis is catalyzed by the enzyme histidine-decarboxylase which converts histidine to histamine. To date two postsynaptic receptors (H1R and H2R) and one presynaptic autoreceptor (H3R), which also functions as a presynaptic heteroreceptor on 2

non-histaminergic nerve terminals, were identified in the rodent brain (Passani and Blandina, 2011). A fourth histamine receptor (H4R) was detected in peripheral tissue (Haas et al., 2008). Histamine receptors differ in terms of pharmacology, localization and cellular transduction processes. The H1R is Gprotein coupled and mediates neuronal excitation and stimulates intracellular second messenger systems which are important for molecular correlates of learning and memory formation. H1R activation stimulates phospholipase C, which, in turn, gives rise to IP3 and Ca2+ release from intracellular Ca2+-stores as well as diacylglycerol formation, which in turn activates the proteinkinase G (Dere et al., 2010). H1R activation can also stimulate phospholipase A and, thereafter, arachidonic acid formation (Haas et al., 2008), which was discussed as a possible post-synaptic retrograde messenger responsible for the presynaptic changes associated with synaptic long-term potentiation (Medina and Izquierdo, 1995). H1R-KO mice exhibited learning and memory impairments in various learning and memory tasks including novel object recognition (Dai et al., 2007), novel objects-induced conditioned place preference (Zlomuzica et al., 2008), Barnesmaze (Dai et al., 2007), 8-arm radial-maze (Zlomuzica et al., 2009), spontaneous spatial alternation in the Y-maze (Zlomuzica et al., 2008), and episodic-like memory (Dere et al., 2008). However, there were also reports of normal performance, e.g. in the inhibitory avoidance task (Yanai et al., 1998) and object-place recognition task (Zlomuzica et al., 2008) or improved learning and memory performance in a contextual fear-conditioning task (Dai et al., 2007). In sum, behavioral studies that have been performed so far rather suggest that genetic inactivation of the H1R in the mouse has a detrimental 3

effect on learning and memory performance as assessed by a variety of memory-related tasks. A cellular substrate of hippocampus-dependent learning and memory processes is the generation of new neurons in the adult dentate gyrus, so called adult neurogenesis (Castilla-Ortega et al., 2011). A reasonable number of studies indicated an association between the rate of adult hippocampal neurogenesis and spatial learning performance (e.g. Drapeau et al., 2003; Kempermann and Gage, 2002; Wolf et al., 2006). With some exceptions (e.g. Shors et al., 2002, Hernández-Rabaza et al., 2009) ablation or suppression of hippocampal neurogenesis in adult mice results in impaired spatial learning and memory (e.g. Deng et al., 2009; Dupret et al., 2008; Snyder et al., 2005). The role of endogenous histamine for adult neurogenesis has not been resolved yet. However, it has been shown in-vitro that histamine induces proliferation of neural progenitors via H2R (Molina-Hernández and Velasco, 2008) and that the neuronal differentiation of progenitor cells is mediated by H1R activation (Molina-Hernández and Velasco, 2008; Rodríguez-Martínez et al., 2012). Histamine-loaded micro-particles facilitated neuronal differentiation both, in slice cultures and in-vivo (Bernardino et al., 2012). Furthermore, it has recently been shown that the H1R antagonist chlorpheniramine prevents histamine induced neuronal differentiation (Molina-Hernández et al., 2013). Thus, there is strong evidence that the elevated histamine levels observed during development play an important role in cerebrocortical neurogenesis while its role in adult neurogenesis has not been sufficiently explored yet. In order to test the generality of the learning and memory impairments observed in the H1R-KO mice, the aim of the present study was to test whether the H1R-KO mice would 4

also show changes in different types of negatively reinforced spatial learning and memory tasks using the water-maze paradigm. This test allows the measurement of different cognitive functions including sensory-motor learning, spatial navigation, spatial learning, and cognitive flexibility (as measured with the reversal test) using the same experimental setup allowing specific conclusions on whether one or more of these functions are selectively disturbed in H1R KO mice. In addition, adult hippocampal neurogenesis was investigated as possible cellular substrate of hippocampus-dependent spatial learning and memory.

2. Experimental Procedures 2.1. Animals, breeding and housing Homozygous H1R-KO mice were delivered from the Riken Research Center for Allergy and Immunology in Yokohama, Japan and were maintained at the animal breeding facilities of the University of Düsseldorf. The generation of H1R-KO mice and the absence of H1R-antagonist binding in the brain of homozygous H1R-KO mice had been described elsewhere (Inoue et al., 1996). The H1R-KO mice founder animals were mated with both male and female C57BL/6JBomTac mice purchased from Taconic/Artemis, Cologne, Germany. The genetic background of the H1R-KO mice was homogenized by backcrossing to the C57BL/6JBomTac strain for at least 10 generations. Heterozygous H1R-KO mice were intercrossed to obtain homozygous H1R-KO and WT mice for behavioral and neurogenesis studies. 10 adult male H1R-KO and 12 WT mice were used for behavioral experiments. Other batches of 9 H1R-KO and 8 WT mice were used for the analysis of adult hippocampal 5

neurogenesis. The animals were housed in Makrolon cages with a 12 h lightdark cycle (lights on from 7 am to 7 pm), had free access to food and water and were maintained under temperature and humidity controlled conditions. All experiments were carried out during the light cycle, between 9 am and 4 pm. Experiments were performed in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC) and all efforts were made to minimize the number of animals used and their suffering. All experiments were announced to the local authority and were approved by the `Animal Welfare Officers' of the Universities of Düsseldorf and Münster.

2.2. Water-maze experiments In order to detect specific changes in negatively reinforced spatial learning in H1R-KO mice, the animal’s performance during different phases of the watermaze learning procedure was assessed. The sequence of tests was as follows: a) cued version, b) place learning, c) spatial probe, d) reversal learning and e) long-term retention (Frisch et al., 2000). For a detailed description of the experimental protocol used see Supplementary Material 1. 2.2.1. Apparatus The water-maze apparatus was a black circular tank with a diameter of 110 cm and a height of 40 cm. It was filled to a depth of 25 cm with water (20 ± 2°C) made opaque white by the addition of 1 l of durable milk. The escape platform was made of transparent Plexiglas, had a diameter of 10 cm and was heightadjustable. The position of the water-maze in the testing room was invariant and the environment contained several stationary distal and proximal visual and 6

acoustic extramaze cues, which might be used for spatial orientation and navigation. The water-maze was virtually divided into 4 quadrants. 2.3. Analysis of adult hippocampal neurogenesis 2.3.1. BrdU injection and tissue preparation At 11 weeks of age H1R-KO mice and WT mice received an intraperitoneal injection twice daily for three consecutive days with 50 mg/kg 5-bromo-2deoxyuridine (BrdU; B5002, Sigma-Aldrich, Munich, Germany) dissolved in 0.9 % NaCl. Four weeks later, mice were deeply anesthetized and perfused transcardially with 0.9 % NaCl, followed by 4 % PBS buffered paraformaldehyde (PFA). Subsequently, brains were fixed 24 h in 4 % PFA at 4°C, washed in tap water and incubated in 30 % sucrose (dissolved in PBS). Coronal sections (40 μm thick) were cut throughout the entire dentate gyrus on a cryostat (Leica). Sections were stored in cryoprotectant (containing 25 % ethylene glycol and 25 % glycerol diluted in PBS) at –20°C until use. 2.3.2. BrdU staining and quantification of BrdU positive cells To assess a possible reduction of newborn neurons, cryosections were stained for BrdU as initially described with slight modifications (Herring et al., 2009). Every sixth cryosection was rinsed extensively in PBS before blocking endogenous peroxidases via incubation in 0.6 % hydrogen peroxide diluted in PBS for 30 min at room temperature (RT). Afterwards, sections were incubated in 2 M HCl for 30 min at 37°C to denature DNA and neutralized in 0.1 M borax for 15 min at RT. After rinsing in PBS, sections were blocked in PBS plus, containing 0.5 % Triton-X100 and 1.5 % normal rabbit serum for 60 min at RT. Sections were incubated overnight at 4°C in monoclonal rat anti-BrdU (1:500, 7

OBT-0030, AbDserotec) diluted in PBS plus. The next day, sections were rinsed several times in PBS and incubated for 60 min at RT in biotinylated rabbit antirat antibody (1:200, BA-4001, Vector Laboratories). After rinsing in PBS, sections were incubated for 30 min at RT in VectastainABC (Vector Laboratories). Sections were rinsed again and peroxidase immunolabeling was conducted by incubating slices for 8 min at RT with DAB (Vector Laboratories). BrdU positive nuclei were counted in the granular cell layer (GCL) and subgranular zone (SGZ) which was defined as two cell layer wide area between the GCL and the hilus in all stained sections using an inverse light microscope (Olympus, IX-81). To estimate the total number of surviving newborn cells for the entire dentate gyrus, the sum of all counted cells for each region was multiplied by six. To avoid double counting, nuclei at the uppermost focal plane of each slice were excluded from analysis. On average 11 sections of each animal were stained and counted. This number did not differ between the genotypes. 2.3.3. BrdU triple labeling and phenotypic quantification of newborn neurons and astrocytes In order to determine the rate of newborn cells that developed into neurons or astrocytes, every twelfth cryosection was stained for BrdU, NeuN, and GFAP as described before with slight modifications (Herring et al., 2009). First, sections were rinsed extensively in PBS, followed by incubation in 50 % formamide/2x SSC solution for 2 h at 65°C and subsequent treatment with 2x SSC for 15 min at RT. Afterwards, sections were incubated in 2 M HCl for 30 min at 37°C and neutralized in 0.1 M borax for 15 min at RT. After rinsing in PBS, slices were blocked in PBS plus containing 0.3 % Triton-X100 and 3 % normal goat serum 8

for 60 min at RT. Sections were incubated overnight in a primary antibody mix diluted in PBS plus (rat anti-BrdU, 1:300, OBT-0030, AbDserotec; mouse antiNeuN, 1:200, MAB377, Millipore; rabbit anti-GFAP, 1:500, Z0334, DAKO) at 4°C. Then, slices were again extensively rinsed in PBS and incubated in a secondary antibody mix diluted in PBS plus (Alexa Fluor A488 anti-rat, 1:300; Alexa Fluor A546 anti-mouse, 1:300; Alexa Fluor A633 anti-rabbit, 1:300, Life Technologies) for 60 min at RT, followed by extensive washing in PBS and coverslipped using fluorescent mounting medium (DAKO). Sections were stored at 4°C in the dark until analysis using a confocal laser-scanning microscope (Zeiss, LSM 710). For each animal 50 randomly chosen BrdU positive cells were analyzed for co-localization with either NeuN or GFAP and the percentages of newborn neurons and astroglia were calculated. In addition, these values were multiplied with the total number of surviving cells in order to calculate the total number of surviving neurons and astrocytes. 2.4. Statistical analysis Data are presented as group means (± SEM). Cued version, place and reversal learning and long-term retention data were analyzed by means of repeated measures ANOVAs. Spatial probe data were analyzed with t-tests for paired and unpaired data. For the analysis of adult hippocampal neurogenesis cell numbers of WT and KO mice were compared with t-tests for unpaired data. Pvalues given are two-tailed and were considered to be significant when p-values < 0.05 were found. In order to compensate for the increase in the probability of finding a significant p-value due to multiple comparisons using the same data set, only p-values < 0.01 were considered to be significant in the spatial probe test. Tests were calculated using the software package SPSS (version 20, IBM). 9

3. Results 3.1.1. Water-maze (Cued version) Across the 3 sessions the animals of both groups showed significant reductions in the time needed to swim to the cued escape platform (escape latency, main effect of session: F(2,40)=72.49; p0.05; Fig. 1C). However, while there was no significant main effect of genotype evident (p>0.05), the sessions x genotype interaction was significant for the swim speed parameter (F(2,40)=4.62; p