Mol Neurobiol DOI 10.1007/s12035-017-0659-3

Fgf9Y162C Mutation Alters Information Processing and Social Memory in Mice Lillian Garrett 1,2 & Lore Becker 2,3 & Jan Rozman 2,3,4 & Oliver Puk 1,2,5 & Tobias Stoeger 2,6 & Ali Önder Yildirim 2,6 & Alexander Bohla 2,6,7 & Oliver Eickelberg 2,6 & Wolfgang Hans 2,3 & Cornelia Prehn 2,3 & Jerzy Adamski 3,8 & Thomas Klopstock 9,10,11 & Ildikó Rácz 12 & Andreas Zimmer 12 & Martin Klingenspor 13 & Helmut Fuchs 2,3 & Valerie Gailus-Durner 2,3 & Wolfgang Wurst 1,10,11,14 & Martin Hrabě de Angelis 2,3,4,8 & Jochen Graw 1,2 & Sabine M. Hölter 1,2

Received: 16 January 2017 / Accepted: 14 June 2017 # Springer Science+Business Media, LLC 2017

Abstract In neuropsychiatric diseases, such as major depression and anxiety, pathogenic vulnerability is partially dictated by a genetic predisposition. The search continues to define this genetic susceptibility and establish new genetic elements as potential therapeutic targets. The fibroblast growth factors (FGFs) could be interesting in this regard. This family of signaling molecules plays important roles in development while also functioning within the adult. This includes effects on aspects of brain function such as neurogenesis and synapse

formation. Of this family, Fgf9 is expressed in the adult brain, but its functional role is less well defined. In this study, we examined the role of Fgf9 in different brain functions by analyzing the behavior of Fgf9Y162C mutant mice, an Fgf9 allele without the confounding systemic effects of other Fgf9 genetic models. Here, we show that this mutation caused altered locomotor and exploratory reactivity to novel, mildly stressful environments. In addition, mutants showed heightened acoustic startle reactivity as well as impaired social discrimination

Electronic supplementary material The online version of this article (doi:10.1007/s12035-017-0659-3) contains supplementary material, which is available to authorized users. * Sabine M. Hölter [email protected]

8

Experimental Genetics, Faculty of Life and Food Sciences Weihenstephan, Technische Universität München, Freising-Weihenstephan, Munich, Germany

1

Institutes of Developmental Genetics, Helmholtz Zentrum München, Neuherberg, Germany

9

Department of Neurology, Friedrich-Baur Institute, Klinikum der Ludwig-Maximilians-Universität München, Munich, Germany

2

German Mouse Clinic, Helmholtz Zentrum München, Neuherberg, Germany

10

Deutsches Zentrum für Neurodegenerative Erkrankungen e. V. (DZNE), Munich, Germany

3

Experimental Genetics, Helmholtz Zentrum München, Neuherberg, Germany

11

4

German Center for Diabetes Research (DZD), Neuherberg, Germany

Munich Cluster for Systems Neurology (SyNergy), Adolf-Butenandt-Institut, Ludwig-Maximilians-Universität München, Munich, Germany

5

Present address: CeGaT GmbH, Paul-Ehrlich-Str. 23, 72076 Tübingen, Germany

12

Institute of Molecular Psychiatry, Medical Faculty, Universität Bonn, Bonn, Germany

13

Molecular Nutritional Medicine, Faculty of Life and Food Sciences Weihenstephan, Technische Universität München, Munich, Germany

14

Developmental Genetics, Faculty of Life and Food Sciences Weihenstephan, Technische Universität München, Freising-Weihenstephan, Munich, Germany

6

7

Comprehensive Pneumology Center, Institute of Lung Biology and Disease, Helmholtz Zentrum München, The German Center for Lung Research (DZL), German Research Center for Environmental Health, Neuherberg, Germany Present address: Satorius Stedim Biotech, Aubagne, France

Mol Neurobiol

memory. Notably, there was a substantial decrease in the level of adult olfactory bulb neurogenesis with no difference in hippocampal neurogenesis. Collectively, our findings indicate a role for the Fgf9Y162C mutation in information processing and perception of aversive situations as well as in social memory. Thus, genetic alterations in Fgf9 could increase vulnerability to developing neuropsychiatric disease, and we propose the Fgf9Y162C mutant mice as a valuable tool to study the predictive etiological aspects. Keywords Fibroblast growth factor 9 . Neuropsychiatric disease . Mice . Brain . Adult neurogenesis

Introduction It is increasingly appreciated that obstruction of developmental pathways plays a pathogenic role in neuropsychiatric disease. Furthermore, there is a growing recognition that clinically distinct diseases, from major depression to schizophrenia, may have a partially shared genetic provenance [1]. Targeting these convergent pathways is thus a potential therapeutic strategy for the resolution of clinical symptoms. In this context, a family of signaling molecules, the fibroblast growth factors (FGFs), is of interest. Consisting of 22 members, they play important roles in various processes including cell fate specification, neurogenesis, and synapse formation in nervous system development as well as being functional in the adult brain [2–4]. Evidence for a role played by these FGFs in neuropsychiatric disease has been garnered from preclinical genetic models. For example, mutations in various FGF family member genes in mice can lead to features analogous to depression [5, FGF22], anxiety [6, 7, FGF2, FGF8], autism [8, FGF17], and epilepsy [9, FGF7]. Recently, a disease-relevant role for FGF9 in major depression was proposed. This was supported by clinical data showing increased expression of FGF9 in hippocampus and frontal cortex from depressed patients [10, 11]. Preclinical analysis further indicated that chronic stress increases hippocampal FGF9 levels, and that administration of FGF9 leads to depression- and anxiety-related behavior. Moreover, knockdown of hippocampal FGF9 was anxiolytic in rats [11]. Blocking the actions of FGF9 was therefore proffered as a novel therapeutic strategy for depression treatment [11]. Nevertheless, a comprehensive model of FGF9 mechanism of action in the brain is not complete nor is it known whether this gene has relevance for other neuropsychiatric disorders. FGF9 is also peripherally expressed in other tissues and organs. It is therefore not surprising that existing preclinical models with mutations in and alterations of the level of FGF9 indicate a role for this molecule outside the central nervous system. Mice with the null phenotype, the Fgf9null mice, have a significantly smaller lung than that of wild types [12]. This

phenotype was similar to the previously reported Bclassical^ knockout, the Fgf9−/− mice [13]. The homozygous Fgf9−/− mutant mice also exhibit a male-to-female sex reversal [14]. This latter feature seems to be unique to the knockout mice, as no sex ratio difference has been observed in the spontaneous Eks mutant mice [15]. The underlying mutation in the Eks mice affects the Fgf9 gene (Asn143Thr), and these mice primarily suffer from elbow and knee joint synostosis (Fgf9Eks or Fgf9Asn143Thr) and premature fusion of cranial sutures [16]. We have reported previously a novel neomorphic allele of Fgf9 in the mouse, Fgf9Y162C. The Fgf9Y162C allele shows an ENU-induced A to G transition resulting in a tyrosine-tocysteine substitution at amino acid 162. The Y162C mutation causes reduced axial eye length, decreased lens size, and impaired visual acuity (over 40% reduction). Nevertheless, no other overt systemic abnormalities outside the eye were detected in these mice [17]. In a bid to further understand the nuances of Fgf9 gene function in the central nervous system, without the confounding physical aberrations that characterize existing FGF9 genetic models, we exposed these Fgf9Y162C mutant mice to a series of emotional and cognitive behavioral and sensorimotor tests. Furthermore, to provide additional insight into the effect of this point mutation on brain function—that may have translational relevance for human neuropsychiatric disorders—we performed concomitant semi-quantitative measures of adult neurogenesis, the birth, and assimilation of new neurons, in the olfactory bulbs and hippocampus of these mice. FGF9 can increase the number of adult subventricular zone neural progenitor cells in vitro by enhancing proliferation and inhibiting astrocyte differentiation [18]. Thus, we hypothesized that alterations in adult neurogenesis could underlie certain behavioral changes in these mutant mice. The outcome of this analysis reveals the Fgf9Y162C mutant line as a potential novel model to study the role of this gene in the shared etiologies of neuropsychiatric diseases.

Materials and Methods Animals Mice were kept under specific pathogen-free conditions at the Helmholtz Center Munich. The use of animals was in accordance with the German Law of Animal Protection, the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and the tenets of the Declaration of Helsinki. All tests performed and described here were approved for the ethical treatment of animals by the responsible authority of the Regierung von Oberbayern (Government of Upper Bavaria). Mice were kept in a 12/12-h dark-light cycle and provided ad libitum standard chow and water. The Fgf9Y162C mutant line [17] is already available to the research

Mol Neurobiol

community via the European Mouse Mutant Archive (EMMA; https://www.infrafrontier.eu; order no. 05137). Two cohorts of mice were analyzed as part of this study (see Fig. 1): from cohort 1 (male and female, wild types (Fgf9WT) and homozygous mutant (Fgf9Y162C) mice derived from timed heterozygous x heterozygous matings, all born within 1 week), group 1 with 8–10 mice were tested in the open field at the age of 9 weeks, grip strength and rotarod at 9– 10 weeks, prepulse inhibition of the acoustic startle at 11 weeks, hot plate at 12 weeks, and adult neurogenesis assessment of brain tissue from a subset of male mice at 17 weeks as follows: olfactory bulb (OB) doublecortin (DCX)+ cell analysis: n = 5 male Fgf9WT, n = 4 male Fgf9Y162C, dentate gyrus (DG) DCX+ cell analysis: n = 4 male Fgf9WT, n = 4 male Fgf9Y162C, DG proliferating cell nuclear antigen (PCNA)+ cell analysis: n = 4 male Fgf9WT, n = 4 male Fgf9Y162C, subventricular zone (SVZ) PCNA+ cell analysis: n = 5 male Fgf9WT, n = 4 male Fgf9Y162C, rostral migratory stream (RMS) PCNA+ cell analysis: n = 5 male Fgf9WT, n = 4 male Fgf9Y162C. Group 2 from cohort 1 with 7–10 mice per group underwent a simple hearing test, the click box, at 9 weeks, locomotor and rearing activity analysis in the PhenoMaster cages at 12–13 weeks. Cohort 2 consisted of an additional 8–11 homozygous mutant animals and corresponding controls of each sex and were used for additional behavioral phenotyping and analysis of plasma corticosterone levels according to the timeline in Fig. 1. Behavior/neurological analysis For the assessment of basic neurological functions, muscle function as well as motor coordination and balance, grip strength, and rotarod performance were measured as Fig. 1 Overview of Fgf9Y162C mutant mice cohorts and controls used in this study, tests performed in sequence and age of mice (in weeks) at time of testing

described previously [19]. Behavioral analysis of spontaneous locomotion, exploration, and anxiety-related behavior were done by open field (mice were 9 weeks of age) and light/dark box (mice were 20 weeks of age) as described before [20, 21]. Tests of spatial working memory in the Y maze (mice were 24 weeks old) and social recognition memory (mice were 32 weeks old) were performed also as described previously [21, 22]. Open Field The open field analysis was carried out as we described previously [20–23]. It consisted of a transparent and infrared light permeable acrylic test arena with a smooth floor (internal measurements 45.5 × 45.5 × 39.5 cm). Illumination levels were set at approx. 150 lx in the corners and 200 lx in the middle of the test arena. Data were recorded and analyzed using the ActiMot system (TSE, Bad Homburg, Germany). Prepulse Inhibition (PPI) of the Acoustic Startle Response Animals were separated based on sex, but not genotype. PPI was assessed automatically using an automated startle apparatus setup (Med Associates Inc., VT, USA) including four identical sound-attenuating cubicles. The protocols were written using the Med Associates BAdvanced Startle^ software. Experiments were carried out between 08:30 and 17:00 h. Background noise was 65 dB, and startle pulses were bursts of white noise (40 msec). A session was initiated with a 5-min acclimation period followed by five presentations of leader startle pulses (110 dB) that were excluded from statistical analysis. Trial types for the PPI included four different prepulse intensities (67, 69, 73, 81 dB); each prepulse

Mol Neurobiol

preceded the startle pulse (110 dB) by a 50-msec interstimulus interval. Each trial type was presented ten times in random order, organized in ten blocks, each trial type occurring once per block. Intertrial intervals varied from 20 to 30 s. This protocol is based on the protocol used in IMPRESS from the International Mouse Phenotyping Consortium (IMPC, see www.mousephenotype.org/impress), adapted to the specifications of our startle equipment. Click Box Test of Hearing Ability The click box was used as a simple test of hearing ability using a protocol described previously [24]. In short, a click box was used to produce a short 20 kHz sound and the Preyer’s reflex is observed and rated on a scale from 0 to 5 where 0 = no reaction and 5 = a strong reaction. The scores are recorded by an observer for each animal and compared. Hot Plate In nociception, mice were screened using a hot plate assay. A mouse was placed on a 28-cm diameter metal surface maintained at 52 + 0.2 °C surrounded by a 20-cm high Plexiglas wall (TSE, Bad Homburg, Germany). Mice remained for 30 s on the plate or until they performed one of three behaviors regarded as indicative of nociception: hind paw licking, hind paw shake/flutter, or jumping. The latency of the first sign of pain was recorded by an observer and compared. Light/Dark Box The test box was made of Plexiglas and divided into two compartments, connected by a small tunnel (4.5 × 5.6 × 13 cm high). The lit compartment (26.1 × 22.6 × 26 cm high) was made of transparent Plexiglas and was illuminated by cold light with an intensity of 650 lx in the middle; the dark compartment (14 × 22.6 × 26 cm high) was opaque, with a lid and not directly illuminated (approx. 5 lx in the center). The mouse was placed in the center of the dark compartment facing the hind wall and allowed to freely explore the apparatus for 5 min. Data were recorded and analyzed using the ActiMot infrared beam break system (TSE, Bad Homburg, Germany). Y Maze Spontaneous alternations were assessed using the Y maze, which was made of opaque light gray PVC and had three identical arms (30 × 5 × 15 cm) placed at 120° from each other; illumination in the center of the maze was 100 lx. Each mouse was placed at the end of one arm and allowed to move freely through the maze during a 5-min session. Spontaneous alternations (defined as consecutive entries into

all three arms without repetitions in overlapping triplet sets) were scored. Total numbers of arm entries were collected cumulatively over the 5 min. Spontaneous alternation performance percentage is defined as the ratio of actual (total alternations) to possible alternations (total number of triplets) × 100. When placed in the Y maze, normal mice prefer to explore the least recently visited arm and thus tend to alternate visits between the three arms. To explore the three arms successively, the mouse must maintain an ongoing record of the most recently visited arms and continuously update such records. Therefore, alternation behavior is a measure of spatial working memory.

Social Discrimination The social discrimination procedure consisted of two 4-min exposures of stimulus animals (ovariectomized 129Sv females) to the test animal in a fresh cage to which the test animal had been moved 2 h prior to testing. All stimulus animals are identified using colored nontoxic nonpermanent paint markers on the tail. During the first exposure, the sample phase, one stimulus animal was introduced into the cage and allowed to roam freely with the test animal. The amount of time that the test animal explores (sniffing of the head and body, direct contact) the stimulus was recorded by a trained observer with a hand-held computer. This measure of the sample phase was used as an index of social investigation and social affinity. After a retention interval of 2 h, this stimulus animal was re-exposed to the test animal together with an additional, previously not presented stimulus animal. A separate Bfamiliar^ and Bunfamiliar^ stimulus animal was assigned to each test animal. The duration of investigatory behavior of the test animal towards the stimulus animals (familiar and unfamiliar) during this test phase was again recorded by a trained observer with a hand-held computer. A social recognition index was calculated as time spent investigating the unfamiliar stimulus mouse/ time spent investigating both the familiar and unfamiliar stimulus mouse.

Grip Strength The grip strength meter system (Bioseb, Chaville, France) determines the grip strength of the limbs, i.e., muscle strength of a mouse. The device exploits the tendency of a mouse to grasp a horizontal metal grid while being pulled by its tail. During the trial setup, the mouse grasps a special adjustable grid mounted on a force sensor. The mouse can catch the grid with either two or four paws. Three trials were undertaken for each mouse and measurement within 1 min. The mean values are used to represent the grip strength of a mouse.

Mol Neurobiol

Rotarod The rotarod (Bioseb, Chaville, France) was used to measure fore limb and hind limb motor coordination, balance, and motor learning ability. The machine was set up in an environment with minimal stimuli such as noise and movement. The rotarod device is equipped with a computer controlled motordriven rotating rod. The unit consists of a rotating spindle and five individual lanes, one for each mouse. In general, the mouse is placed perpendicular to the axis of rotation, with head facing the direction of the rotation. All mice were placed on the rotarod at an accelerating speed from 4 to 40 rpm for 300 s with 15 min between each trial. In motor coordination testing, mice were given four trials at the accelerating speed at 1 day. The mean latency to fall off the rotarod during the trials was recorded and used in subsequent analysis. Before the start of the first trial, mice were weighed.

(Leica, Bensheim) into 8-μm-thick sagittal sections and stored at 4 °C. A one-in-eight series of sections was taken for analysis. The slides from each series were coded to ensure that the observer was blind to the experimental group until analysis. Immunostaining

For the evaluation of home cage locomotor and exploratory activity, single mice were kept in respirometry cages (PhenoMaster System, TSE Systems, Germany) [25]. The measurement started at 1 pm (CET) after a 2-hour adaptation to the cages and continued until 10 am the next morning. The setup allowed the analysis of locomotor activity of individual mice every 20 min resulting in 63 readings per individual and trial. Two infrared light beam frames allowed the monitoring of physical activity (lower frame: distance traveled per 20 min, upper frame: number of rearings per 20 min).

For immunostaining of doublecortin (DCX) and proliferating cell nuclear antigen (PCNA), an avidin-biotin complex (ABC) method like that employed previously [28, 29] was used and adapted for staining paraffin embedded tissue on slides. The adaptations necessary included a heat-induced antigen retrieval step where slides were heated in 0.01 M citric acid (pH 6.0) to 100 °C twice for 5 min with a 10-min cooling period in between. Furthermore, quenching of endogenous peroxidase with 0.3% H2O2 was carried out for 5 min. Primary antibody incubations took place over night at 4 °C and secondary antibody as well as ABC (VECTASTAIN Elite ABC HRP Kit PK-6100, VECTOR LABORATORIES, INC., Burlingame, USA) incubations took place for 1 h on the second day. A primary rabbit polyclonal anti-DCX antibody (1:200, Catalog #: Ab18723, Abcam) was used in this protocol with a biotinylated goat anti-rabbit IgG (1:300; Biotin-SP AffiniPure Goat Anti-Rabbit IgG, Jackson ImmunoResearch Inc., USA), and a primary monoclonal mouse anti-PCNA antibody (1:200, PC10, Catalog #: Ab29, Abcam) was used with a biotinylated rabbit anti-mouse IgG (1:300, Biotin-SP AffiniPure Rabbit Anti-Mouse IgG, Jackson ImmunoResearch Inc., USA) both with 3,3′-diaminobenzidine (DAB) as the chromogen. Negative controls, with omission of the primary antibodies, revealed no positive staining (see Supplemental Fig. 1).

Steroid Metabolism

Quantification of DCX-and PCNA Positive Neurons

Blood samples were collected in the morning by retrobulbar puncture under isoflurane anesthesia. Steroid hormone corticosterone concentrations were quantified out of 20 μL EDTA plasma using a validated high-throughput LC-MS/MS method (HPLC: Shimadzu Prominence, Shimadzu Deutschland GmbH, Duisburg, Germany; MS/MS: 4000 QTRAP, Sciex Deutschland GmbH, Darmstadt, Germany). Sample preparation and LC-MS/MS measurements have been described in detail [26, 27]. Concentrations are reported in nmol/L and have been calculated using stable isotope-labeled internal standards and reference standard curves. The sensitivity of the test was 2.8 nmol/L for corticosterone.

Quantification of the number of DCX+ and PCNA+ cells was performed using a Zeiss Axioplan2 microscope equipped with StereoInvestigator image analysis software (MicroBrightField Inc., Williston, VT, USA). Using the ×10 objective, the dorsal hippocampal dentate gyrus and the granular cell layer of the main olfactory bulbs were contoured in one hemisphere using the software. The olfactory bulb granular cell layer was analyzed between the lateral boundaries of 0.60 and 0.96 mm, the dorsal dentate gyrus between 0.96 and 1.44 mm according to the mouse brain atlas of Paxinos and Franklin [30]. Subventricular zone and rostral migratory stream and the subgranular zone of the dorsal hippocampal dentate gyrus (defined as a two-cell width at the base of the granular cell layer) were traced and contoured using the software for analysis of PCNA immunopositive cells between 0.96 and 1.44 mm. All cells within the contours were counted by scanning through the tissue in the x-y plane using the software and a ×20 objective. DCX immunopositive cells were only counted when clear cell bodies were visible or clear elliptical stain

Locomotor and Exploratory Activity in PhenoMaster Cages

Tissue Preparation Mice were sacrificed with CO2 and after resection brains were fixed in 4% buffered formalin and embedded in paraffin for histological examination. Brains from a subset of male animals from each group were then sectioned on a microtome

Mol Neurobiol

was evident in the case of PCNA. For OB DCX, eight sections were counted, eight sections for DG DCX, eight sections for DG PCNA, and six sections for each SVZ and RMS PCNA. The total cell counts were multiplied by section sampling interval (8) and the section thickness (8 μm) to estimate the total number of immunopositive cells within these regions of interest. Statistical Analysis Numerical analyses were performed using GraphPad Prism version 7.03 for Windows (GraphPad Software, La Jolla, California, USA, www.graphpad.com) and R version 3.0.2 [31]. For continuous data meeting the assumption of normality, a two-way ANOVA was performed to test genotype-sex interaction effects (all two-way ANOVA results are shown in Table 1 for open field, light/dark box, grip strength, corticosterone, prepulse inhibition, hot plate, and social discrimination). When significant interactions were detected (for corticosterone), a post hoc Bonferroni’s test was used to determine differences between groups. For tests with repeated measures (acoustic startle, rotarod, PhenoMaster home cage activity), we used a repeated measures (RM) ANOVA with decibel level, trial and time of day, respectively, as within-subject factors and genotype as between-subject factor (sexes were pooled for this analysis). When significant interactions were detected (in acoustic startle and PhenoMaster home cage activity), a post-hoc Sidak's test was performed at individual decibel and time of day levels. When two groups were compared (DCX+ and PCNA+ cell analysis), an unpaired Student’s t test was used. When the assumption of normality was not met, a Kruskal-Wallis ANOVA on ranks (click box test) was performed. For all tests, a P value <0.05 was used as level of significance and data are presented as means + SEM. A correction for multiple testing of the various parameters was not performed.

Results Fgf9Y162C Confers Altered Locomotor and Startle Reactivity In the open field, a 20-min test of locomotor activity, exploration, and anxiety-related behavior, the Fgf9Y162C mutant mice showed decreased forward locomotor activity at the age of 9 weeks, evident in the peripheral zone of the arena without significant differences in activity in the whole arena (two-way ANOVA genotype effect periphery: F(1,31) = 4.43, p = 0.04, Fig. 2a; whole arena: genotype effect F(1,31) = 2.45, p = 0.13, data not shown). There were no clear differences in rearing activity as a measure of exploration or in movement velocity at this age (see Table 1). However, in the light/dark

box, another test of anxiety-related behavior that exploits a rodent’s natural aversion to open brightly lit spaces, there was a strong decrease in total distance traveled and movement velocity in the dark box relative to wild-type mice at the older age of 20 weeks (distance traveled: two-way ANOVA genotype effect: F(1,34) = 18.53, p = 0.0001, sex x genotype interaction effect: F(1,34) = 0.00, p = 0.97; velocity in dark box genotype effect: F(1,34) = 19.15, p = 0.0001, sex x genotype interaction effect: F(1,34) = 0.04, p = 0.84, Fig. 2b, c). This significant locomotor decrease was manifest in each of the chambers: light and dark boxes and connecting tunnel (see Table 1). In addition, mutant mice engaged in significantly less total rearing activity during this 5-min test (two-way ANOVA genotype effect F(1,34) = 4.62, p = 0.04, sex x genotype interaction effect F(1,34) = 0.01, p = 0.91; Fig. 2d). The decrease in locomotor activity in the light/dark box and, to a lesser extent, in the open field could be due partially to impairments in motor ability. Thus, to address this possibility, the Fgf9Y162C mutant mice were also assessed in the accelerating rotarod, as a measure of motor coordination, motor learning, as well as balance. Reduced locomotor activity in these mice was ostensibly not due to alterations in motor coordination as there were no clear genotype-related differences in the latency to fall off the rotarod (RM ANOVA: trial x genotype interaction effect: F(2,30) = 2.42, p = 0.10, Fig. 2e). Furthermore, there were no marked effects of the mutation on grip strength (Table 1), to indicate that the locomotor effects were consequent to a loss of muscular strength. We analyzed the locomotor (distance traveled) and exploratory (rearing) activity by the Fgf9Y162C mutant mice in the home cage environment of the PhenoMaster cage over a 21-h period. During the analysis, the mice were given sufficient time to habituate to the novel home cage. It is expected that a pervasive decrease in activity due to motor disability would be detected here, without the confounding influence of anxiety on the animal’s response. Moreover, the continuous analysis over a protracted period permits the detection of differences in circadian rhythm. Mice were placed singly into these novel home cages and activity was recorded. Interestingly, during the habituation time-bin, i.e., the first 3 h after transfer of the animals to these cages, the Fgf9Y162C mutant mice exhibited a sustained hyperactivity in both distance traveled and rearing activity without clear differences between the genotypes for the remainder of the test (habituation phase, total distance RM ANOVA: genotype x time effect: F(8,128) = 3.12, p = 0.003, Table 1, post hoc Sidak’s test: 60 min: t(128) = 3.96, p < 0.01, 80 min: t(128) = 5.06, p < 0.0001, 100 min: t(128) = 3.12, p < 0.05, Fig. 3 a, b; habituation phase, rearing activity RM ANOVA: genotype x time effect: F(8,128) = 6.03, p < 0.001, Table 1, post hoc Sidak’s test: 20 min: t(12) = 6.35, p < 0.0001, 40 min: t(128) = 5.00, p < 0.0001, 60 min: t(128) = 5.17, p < 0.0001, 80 min: t(128) = 5.70, p < 0.0001, 100 min: t(128) = 3.11,

Recognition index Investigation time (s)

Social discrimination

Second response time (s) Y maze Spontaneous alternations (%) Alternate arm returns (%) 1,34

0.000 1,34 1,34

1,34

0.007

7.82 0.61

1,31

1,31

0.002 0.003

1,31 1,31

1,34 1,34 1,34

1,34 1,34 1,34 1,34 1,34 1,34

2.85 0.73

0.84 1.28 19.54

Entries into light box (#) Time in light box (%) Velocity light box (cm/s)

Grip strength 2.85 0.73 Hot plate First response time (s)

18.53 19.15 22.33 13.50 10.68 4.62

1,31

1,31 1,31 1,31 1,31 1,31

2.45 0.25 2.28 9.75 10.76 0.58

1,31

4.43

Total distance (cm) Velocity dark box (cm/s) Distance in dark box (cm) Distance in light box (cm) Distance in tunnel (cm) Rearing (#)

Center entries (#) Light/dark box

Distance moved periphery (cm) Distance moved total (cm) Rearing (#) Velocity (cm/s) Center time (%) Center velocity (cm/s)

Open field

0.008 0.44

0.99

0.93

0.95

0.97

0.10 0.40

0.37 0.27 <0.0001

0.0001 0.0001 <0.0001 0.0008 0.003 0.04

0.45

0.13 0.62 0.14 0.004 0.003

0.04

2.33 1.50

0.26

0.73

0.10

0.25

1.02 0.81

0.07 3.04 0.80

1.68 0.97 0.20 1.53 4.78 1.14

0.17

1.51 7.48 1.68 0.88 2.45

2.62

F

p

F

df

Sex

Genotype

1,34 1,34

1,34

1,34

1,31

1,31

1,31 1,31

1,34 1,34 1,34

1,34 1,34 1,34 1,34 1,34 1,34

1,31

1,31 1,31 1,31 1,31 1,31

1,31

df

0.14 0.23

0.61

0.40

0.75

0.62

0.32 0.38

0.79 0.09 0.38

0.20 0.33 0.66 0.22 0.04 0.29

0.69

0.23 0.01 0.20 0.35 0.13

0.12

p

0.74 2.25

1.16

1.01

0.80

0.69

0.30 0.05

0.44 0.39 0.05

0.00 0.04 0.00 0.17 0.63 0.01

1.10

2.17 2.02 2.09 0.09 1.25

1.64

F

Genotype x Sex

Summary table of two-way ANOVA and repeated measures (RM) ANOVA results from tests performed on male and female Fgf9WT and Fgf9Y162C mice

Two-way ANOVA results with genotype and sex as factors

Table 1

1,34 1,34

1,34

1,34

1,31

1,31

1,31 1,31

1,34 1,34 1,34

1,34 1,34 1,34 1,34 1,34 1,34

1,31

1,31 1,31 1,31 1,31 1,31

1,31

df

0.40 0.14

0.29

0.32

0.38

0.41

0.59 0.82

0.51 0.54 0.70

0.97 0.84 0.99 0.68 0.43 0.91

0.30

0.15 0.17 0.16 0.77 0.27

0.21

p

Mol Neurobiol

6.73 4.33 3.35 0.08 0.02 1.36 1.22 0.67 0.92 0.69 5.20 3.33 3.22

73 dB 1 (%) 81 dB 1 (%) Global 1 (%) 67 dB 2 (%)

69 dB 2 (%) 73 dB 2 (%)

81 dB 2 (%) Global 2 (%) 67 dB 1 + 2 pooled (%) 69 dB 1 + 2 pooled (%) 73 dB 1 + 2 pooled (%)

81 dB 1 + 2 pooled (%) Global 1 + 2 pooled (%)

ASR 1 (arbitrary units) ASR2 (arbitrary units)

Locomotion–habituation (cm) Rearing–habituation (#) Acoustic startle

df 1,16 1,16

df 1, 202 1, 258

Genotype F 2.31 37.49

1,32

1,65 1,65

1,34 1,34 1,65 1,65 1,65

1,34 1,34

1,27 1,27 1,27 1,34

1,27 1,27

F 3.87 7.24

Corticosterone Corticosterone 6.27 Repeated measures ANOVA results (m + f pooled) PhenoMaster cages Genotype

0.92 0.86

67 dB 1 (%) 69 dB 1 (%)

PPI

p 0.13 0

p 0.07 0.02

0.02

0.07 0.08

0.27 0.42 0.34 0.41 0.03

0.89 0.25

0.01 0.05 0.08 0.78

0.35 0.36

dB F 607.86 841.06

F 66.18 65.97

Time

77.05

0.52 0.42

0.36 1.90 1.41 2.77 0.02

4.49 0.88

0.88 2.10 0.20 1.21

0.19 0.11

F

p

F

df

Sex

Genotype

Two-way ANOVA results with genotype and sex as factors

Table 1 (continued)

df 1, 202 1, 258

df 8, 128 8, 128

1,32

1,65 1,65

1,34 1,34 1,65 1,65 1,65

1,34 1,34

1,27 1,27 1,27 1,34

1,27 1,27

df

p 0 0

p <0.0001 <0.0001

<0.0001

0.47 0.52

0.56 0.18 0.24 0.10 0.90

0.04 0.35

0.36 0.16 0.66 0.28

0.66 0.75

p

df 8, 128 8, 128

1,32

1,65 1,65

1,34 1,34 1,65 1,65 1,65

1,34 1,34

1,27 1,27 1,27 1,34

1,27 1,27

df

Genotype x dB F df 2.48 1, 202 19.53 1, 258

F 3.12 6.03

Genotype x time

7.45

0.69 0.12

0.61 0.27 0.08 0.39 0.40

0.26 0.96

0.18 0.18 0.02 0.14

0.01 0.18

F

Genotype x Sex

p 0.12 0

p 0.003 <0.0001

0.01

0.41 0.73

0.44 0.60 0.78 0.53 0.53

0.61 0.33

0.67 0.67 0.88 0.71

0.92 0.67

p

Mol Neurobiol

Mol Neurobiol Fig. 2 Spontaneous vertical (rearing) and horizontal (distance) locomotor activity and speed in response to novelty and bright light in the open field and light/ dark box by Fgf9Y162C mutant mice and controls. a, b Distance traveled in the open field and light/dark box. c Locomotor velocity in the light/dark box. d Rearing activity in the light/dark box. e Latency to fall from the rotarod. *p < 0.05, ***p < 0.001 two-way ANOVA genotype effect, Fgf9WT mice vs. Fgf9Y162C mutant mice. Data are means + SEM

p < 0.05, Fig. 3c, d). The circadian profile of locomotor variation and sleep/wake cycle did not differ significantly between the genotypes. To ascertain the effect of the Fgf9Y162C missense mutation on sensory function in mice, we examined the thermal sensitivity of the mice in the hot plate test. There were no significant genotype effects on either Fig. 3 Home cage locomotor activity of the Fgf9Y162C mutant mice and controls in the PhenoMaster cages. a, b Locomotor activity over the course of the 21-h testing period and increases were detected during the first 3-h habituation phase. c, d Rearing activity over the course of the 21-h testing period and increases were detected during the first 3-h habituation phase. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 Fgf9Y162C vs. Fgf9WT mice from repeated measures ANOVA with post hoc Sidak’s multiple comparison test. Data are means ± SEM

the first or second reaction to a thermal stimulus (Table 1, data not shown). We also assessed sensorimotor recruitment, behavioral reactivity, and sensorimotor gating using the acoustic startle reflex (Fig. 4a) and prepulse inhibition (Fig. 4b). This test was performed on two independent cohorts of mice under the same conditions; data from the two cohorts were pooled for

Mol Neurobiol Fig. 4 Sensorimotor recruitment and sensorimotor gating responses of the Fgf9Y162C mutant mice and controls. a Acoustic startle reactivity of the Fgf9Y162C mutant mice was increased compared to controls. b % Prepulse inhibition at four prepulse intensities and at global. A small decrease in prepulse inhibition was detected significant at the 73 dB prepulse intensity. *p < 0.05 two-way ANOVA genotype effect, ***p < 0.001, ****p < 0.0001 repeated measures ANOVA with post hoc Sidak’s multiple comparison test, Fgf9Y162C vs. Fgf9WT mice. Data are means + SEM

analysis (results from separate analyses shown in Table 1). The Fgf9Y162C mutant mice demonstrated an increased acoustic startle reactivity that was significant at the higher dB intensities (RM ANOVA: genotype effect: F(1,33) = 10.73, p = 0.003, dB effect: F(7,231) = 599.6, p < 0.0001, genotype x dB interaction effect: F(7,231) = 6.18, p < 0.0001, post hoc Sidak’s test: 90 dB: t(231) = 2.31, p < 0.001, 100 dB: t(231) = 4.30, p < 0.0001, 110 dB: t(231) = 5.32, p < 0.0001, Fig. 4a). In terms of prepulse inhibition, there was a small decrease in the Fgf9Y162C mutant mice compared to controls (Table 1, Fig. 4b). This difference was significant at the 73 dB intensity (two-way ANOVA genotype effect: F(1,65) = 5.20, p = 0.03) with tendencies of significance at the 81 dB intensity (two-way ANOVA genotype effect: F(1,65) = 3.33, p = 0.07) and globally (two-way ANOVA genotype effect: F(1,65) = 3.22, p = 0.08) indicating a deficit in sensorimotor gating and pre-attentive processing in these mice. In the click box test of high frequency hearing ability, there were no differences between the genotypes (Kruskal-Wallis ANOVA on ranks: Kruskal-Wallis statistic = 5.11, p = 0.16), suggesting that differences in acoustic startle and prepulse inhibition were not secondary to changes in hearing ability.

Fgf9Y162C Results in Emotional Alterations and a Sex-Specific Difference in Corticosterone Levels After Anesthesia We assessed the Fgf9Y162C mutant mice for anxiety-related behavior in the open field and light/dark box tests. In the open field, the central, unprotected area is the most anxiogenic and the amount of time spent in the center reflects the anxiety levels of the mouse. The Fgf9Y162C mutant mice were spending a greater percentage of time in the central more aversive zone (two-way ANOVA genotype effect: F(1,31) = 9.75, p = 0.004, Table 1, Fig. 5a). The average speed of movement in the center was significantly decreased (two-way ANOVA genotype effect: F(1,31) = 10.76, p = 0.003, Fig. 5b), but no difference in the number of center entries was detected (two-way ANOVA genotype effect: F(1,31) = 0.58, p = 0.17, Fig. 5c). In terms of anxietyrelated behavior in the light/dark box, there were no clear genotype effects on percentage time spent in the light box (two-way ANOVA genotype effect: F(1,34) = 1.28, p = 0.27, Fig. 5d); the mutant mice did however show a decreased movement velocity within this zone (two-way ANOVA genotype effect: F(1,34) = 19.54, p < 0.0001, Fig. 5e). The number of entries into the lit anxiogenic section was unchanged (two-way ANOVA genotype effect:

Mol Neurobiol

Fig. 5 Unconditioned anxiety-related responses of the Fgf9Y162C mutant mice and controls in the open field and light/dark box. a, b, c Time spent in, velocity and number of entries into the center of the open field. Increases in center time detected with a decrease in movement velocity within this zone and no differences in the number of center entries. **p < 0.01, two-way ANOVA genotype effect, Fgf9Y162C vs. Fgf9WT mice. d, e, f Time in the lit compartment of the light/dark box, movement velocity and number of entries into this zone. No difference in time spent

in or entries into the lit compartment were detected. There were decreases in movement velocity in the lit compartment. ****p < 0.0001 two-way ANOVA genotype effect, Fgf9Y162C vs. Fgf9WT mice. g light and dark box preference time. **p < 0.01 paired t test time in light vs. time in dark box. h Plasma corticosterone levels. **p < 0.01 female Fgf9WT mice vs. female Fgf9Y162C mice, two-way ANOVAwith post hoc Bonferroni’s test. Data are means + SEM

F(1,34) = 0.84, p = 0.37, Fig. 5f). Furthermore, both control and the Fgf9Y162C mutant mice showed a similar preference for the dark chamber compared to the light (Fig. 5g), where there was a significant difference between the amounts of time spent in each chamber.

We also examined the plasma levels of the stress hormone corticosterone in anesthetized mice. Here, we observed a sex x genotype interaction effect where the female mutant mice showed increased plasma corticosterone levels compared to controls with no clear differences in the males (two-way

Fig. 6 Cognitive behavior of the Fgf9Y162C mutant mice and controls. a, b There were no genotype-related differences in spontaneous alternations and alternate arm returns in the Y maze. c, d Recognition index and social investigation time in the social discrimination test. **p < 0.01 two-way ANOVA, genotype effect, Fgf9Y162C vs. Fgf9WT mice. Data are means + SEM

Mol Neurobiol

Fig. 7 Adult neurogenesis of the Fgf9Y162C mutant mice and controls. a, b, c There were no differences in the number of doublecortin (DCX)positive cells in the granular cell layer of the hippocampal dentate gyrus (arrows). N = 4/group. Scale bar = 100 μm. d, e, f There was no difference in the number of proliferating cell nuclear antigen (PCNA)positive cells (arrows) in the subgranular zone (SGZ) of the hippocampal dentate gyrus (DG). N = 4/group. Scale bar = 100 μm. g, h, i There was a significant decrease in the number of DCX-positive cells (arrows) in the

granular layer of the olfactory bulbs. **p < 0.01 Student’s t test, t(7) = 3.52, p = 0.009, Fgf9Y162C (n = 4) vs. Fgf9WT (n = 5) mice. Scale bar = 100 μm. j, k, l There was a significant decrease in the number of rostral migratory stream (RMS) PCNA+ cells (Student’s t test, t(7) = 2.44, p = 0.04) and a tendency (Student’s t test, t(7) = 1.96, p = 0.09) to a decrease in the subventricular zone (SVZ) of the Fgf9Y162C (n = 4) vs. Fgf9WT (n = 5) mice. *p < 0.05 Student’s t test. Scale bar = 100 μm. Data are means + SEM

ANOVA, sex x genotype interaction: F(1,32) = 7.45, p = 0.01, post hoc Bonferroni’s: p < 0.01; Table 1, Fig. 5h).

fate of neural progenitor cells in vitro [18], we hypothesized that FGF9 may play a role in mediating functions, such as learning and memory, that involve adult neurogenesis. Thus, we performed a simple hippocampal test of spontaneous alternations in the Y maze. When tested for spatial working memory ability in the Y maze at 22 weeks, there were no clear genotype effects on the number of spontaneous alternations or alternate arm

Fgf9Y162C Mutation Causes a Deficit in Social Discrimination Memory As FGF9 is expressed in the hippocampus and in its related major projections [32], and FGF9 can alter the

Mol Neurobiol

returns (Table 1, Fig. 6a, b) in the mutant mice. We next examined social behavior and social recognition, the ability to distinguish a familiar from an unfamiliar conspecific, in the social discrimination test. This is a test of short-term olfactory learning and memory in a social context. The Fgf9Y162C mutant mice were examined in this test at the older age of 32 weeks. The mutant mice showed a significantly lower social recognition memory index (two-way ANOVA genotype effect: F(1,34) = 7.82, p = 0.008, Table 1, Fig. 6c) without clear genotype differences in the social investigation time (Table 1, Fig. 6d) suggesting a specific social memory impairment that was not due to compromised olfactory abilities in these mice. Fgf9Y162C Impairs Olfactory Bulb Adult Neurogenesis Without Altering Adult Hippocampal Neurogenesis To determine if the behavioral alterations were associated with modifications of the adult neurogenesis in Fgf9Y162C mouse mutants, the number of DCX+ and PCNA+ markers was quantified as indices of differentiation and proliferation, respectively, in the hippocampus and olfactory bulb/ subventricular zone adult neurogenic zones. DCX expression in the adult brain has also been shown to reflect overall levels of neurogenesis and DCX+ cell quantification can be used as an alternative to BrdU pulse/chase analysis [33]. Quantification showed that there was no significant effect of genotype on the number of DCX+ cells in the dentate gyrus (t(6) = 0.10, p = 0.93, Fig. 7a, b, c) or in the number of PCNA+ cells in the subgranular zone of the dentate gyrus (t(6) = 0.54, p = 0.61, Fig. 7d, e, f). In the granular cell layer of the olfactory bulbs, there was a significant decrease in the number of DCX+ cells in the mutant mice compared to controls (t(7) = 3.52, p = 0.009, Fig. 7g, h, i). Furthermore, there was a significant increase in the number of PCNA+ cells in the rostral migratory stream (t(7) = 2.44, p = 0.04) and a tendency towards an increase in the subventricular zone (t(7) = 1.96, p = 0.09, Fig. 7j, k, l).

Discussion A disease-relevant role for FGF9 in major depression has been proposed recently [11], yet a clear understanding of FGF9 function in the brain is lacking. In a bid to increase this understanding, we exposed these Fgf9Y162C mutant mice to assays that interrogate multiple aspects of brain function. We determined that the Fgf9Y162C mutation causes altered locomotor and anxiety responses to novel stressful environments, hyperactive acoustic startle responsivity, as well as impaired social memory. These behavioral transformations were accompanied by a decrease in adult

olfactory bulb neurogenesis, increased number of proliferative cells in the rostral migratory stream without changes in hippocampal neurogenesis. Mutations in the Fgf9 gene thus produce endophenotypes characteristic of different neuropsychiatric disorders, and the Fgf9Y162C mutant mouse is a potential model to assess shared genetic etiologies related to this gene. Fgf9Y162C Disrupts Information Processing Ability, Anxiety, and Startle Responding The diametrically opposed locomotor alterations conseque nt to the Fgf9 Y 1 6 2 C mutation (increases in PhenoMaster cages and decreases in open field, light/ dark box) indicate that this is not a pure motor phenotype, rather hinting towards altered information processing in response to novel stressful environments. The increased locomotor activity on initial exposure to the PhenoMaster home cage, for example, could denote an inability to accrue accurate information to habituate [34]. With more efficient context processing, environmental information would be assimilated faster and habituation would occur sooner. Furthermore, as part of the International Mouse Phenotyping Consortium (IMPC), a heterozygous Fgf9 knockout mouse displayed a phenotype of locomotor hyperactivity in the open field (http://www.mousephenotype.org; search Fgf9 for parameter key: IMPC_OFD_005_001; [35]). The converse pattern of decreased locomotor activity observed in the light/dark box occurred without clear differences in the classic anxiety-related index of time in light box, and the mutant and control groups showed similar degrees of preference for the dark box, which infers that this is not strictly an anxiety effect. While it must be considered that the impaired vision previously reported in these mice could have affected behavior [17], there is evidence to suggest that visually impaired mice behave similarly to visually intact mice in such dark and light conditions [36]. In sum, FGF9 modulates appropriate locomotor responses to stressful situations. Furthermore, these locomotor effects are unlikely to be secondary to systemic effects of the neomorphic Fgf9Y162C point mutation as it does not produce the neonatal lethal phenotype or sex reversal of the Fgf9−/− knockout mice [14, 17] or the skeletal and lung development defects of the knockout mutants and the Fgf9Eks allele [12, 13, 16]. The Fgf9Y162C mutant mice spent longer percentage duration in the center of the open field, indicating that these animals were less anxious in this environment. As before, it is possible that this effect may be secondary to impaired vision that would potentially mitigate the aversiveness of the exposed central zone of the arena. Nevertheless, it is interesting that there is evidence showing that FGF9 is anxiogenic where

Mol Neurobiol

chronic intracerebroventricular administration of FGF9 in rats leads to increased anxiety-related behavior, while the opposite effect was seen with a lentiviral knockdown of hippocampal FGF9 [11]. Thus, it appears that altered anxiety responding could be a primary effect of FGF9 disruption in these neomorphic Fgf9Y162C mice. Of note in this regard, the increased circulating levels of corticosterone, the glucocorticoid stress hormone, in female mutants, infers that the Fgf9Y162C mutation causes a sex-specific dysregulation of the hypothalamic-pituitary-adrenal stress axis under anesthesia. While isoflurane anesthesia is known to increase corticosterone levels [37], the overall implication of this finding is that these mice exhibit a susceptibility to abnormal stress axis reactivity. Acoustic startle reactivity was also augmented in the Fgf9Y162C mutant mice, and a similar effect was observed in the heterozygous Fgf9 knockout mice phenotyped in the IMPC (http://www.mousephenotype. org; search Fgf9 for parameter key: IMPC_ACS_006_ 001). The startle reflex is an involuntary reaction to a loud auditory stimulus that is analogous in rodents and humans but subject to the emotional state of the individual [38]. Acoustic startle hyperreactivity is a symptom of the anxiety disorder posttraumatic stress disorder (PTSD) in humans and it has been conjectured [39], although recently contested in humans [40] and rodents [41], that a propensity to increased startle could portend a predisposition to PTSD. Moreover, there was a small deficit in PPI detected; a measure of the degree to which a non-startling prestimulus can reduce the startle reflex [38]. The deficit likely reflects a decreased ability to filter relevant information from the surroundings, akin to that implicated in human neuropsychiatric disorders such as PTSD, schizophrenia, and autism [42–44]. As mentioned, there is evidence to suggest a role for FGF9 in the pathogenesis of major depressive disorder (MDD, [10, 11]). Nevertheless, these phenotypes of hyperarousal, seen as increased baseline acoustic startle reactivity and heightened responses to aversive situations such as the altered locomotor reactivity observed in Fgf9Y162C mice in the PhenoMaster cages, are not generally observed in depression [45]. Therefore, it appears that this point mutation in FGF9 can lead to features of both anxiety and depression disorders and may thus be a shared genetic association in patients where these diseases are comorbid. Fgf9Y162C Mice Exhibit Impaired Social Recognition Memory and Olfactory Bulb Neurogenesis It was shown previously that FGF9 can increase the number of adult subventricular zone neural progenitor cells in vitro by increasing proliferation and inhibiting astrocyte differentiation

[18]. Adult neurogenesis is the birth of new neurons from neural progenitor cells in the subgranular zone of the hippocampal dentate gyrus and the subventricular zone along the walls of the lateral ventricles. Cells from the latter region will migrate along the rostral migratory stream to be incorporated into the olfactory bulbs as mature neurons [46, 47]. Decreases in adult neurogenesis can lead to alterations in cognitive, olfactory, and emotion-related behaviors [28]. Given this effect of FGF9 on neural progenitor cells, we hypothesized that this FGF9 point mutation would lead to alterations in behaviors associated with adult neurogenesis [28]. While simple spatial working memory in Y maze was unaffected, the Fgf9Y162C mutation impaired social recognition memory. This type of memory recruits several brain areas including the olfactory system, medial amygdala, lateral septum, and the hippocampus [48]. Furthermore, deficits in social memory have been found in human mood disorder and schizophrenia patients [49]. We have shown previously the involvement of adult neurogenesis in social recognition memory in the DCXCreERT2; DTA model where decreased neurogenesis (both olfactory bulb and hippocampal) reversibly impaired social recognition memory [28]. We here found a substantial decrease in olfactory bulb neurogenesis in the mutant mice without differences in the hippocampus. Moreover, we observed a concomitant increase in the number of proliferative cells in the rostral migratory stream with a pattern of an increase in the subventricular zone. This Fgf9 point mutation thus appears to amplify the pro-proliferative effects of FGF9 and/or slows the differentiation into DCX+ immature neurons. In any case, this decrease may explain the impaired recognition memory; however, additional studies will be required to investigate this connection in more detail. This is, to the best of our knowledge, the first in vivo evidence to show that alterations in FGF9 alter adult neurogenesis, complementing the previous in vitro study [18]. A comprehensive characterization of the role of FGF9 in adult neurogenesis is so far lacking. It has been conjectured, based on inverse correlations in levels in the hippocampus, that FGF9 is a functional antagonist of FGF2, the latter known to play a role in adult neurogenesis [11, 50]. Thus, it will be interesting to assess whether there is a disequilibrium in these neurotrophic factors that occurs in these mutant mice and if the effect of FGF9 on adult neurogenesis is mediated by its effects on FGF2. The nature of the dysregulation in Fgf9 in these mice is, yet, incompletely elucidated, but it is likely that this neomorphic Fgf9 mutation possesses a novel molecular function that will require additional characterization. In any case, our data show that Fgf9Y162C mutant mice have reliable social discrimination performance deficits and changes in locomotor activity as well as adult neurogenesis alterations that are not likely due to non-associative factors of dysmorphological and respiratory (unpublished observation) changes observed in other Fgf9 alleles.

Mol Neurobiol

Fgf9Y162C as a Model to Study Shared Etiologies of Neuropsychiatric Disease There is compelling evidence from both human and rodent studies suggesting a role for disrupted FGF9 function in major depressive disorder and anxiety [10, 11]. Some of the phenotypes presented and discussed here for the Fgf9Y162C mutant mice tally with this disease association (anxiety and locomotor changes). Nevertheless, this point mutation produces other brain phenotypes not exclusively emblematic of depression or anxiety and akin to that described in, for example, schizophrenia or autism. This diverse and complex phenotype likely reflects the extensive biological pleiotropy of this gene and highlights how individual genetic associations can be shared across multiple psychiatric disorders. In a new era of precision medicine, where nosology and diagnostic criteria could be based on the cause rather than the symptoms of a disease, we propose the Fgf9Y162C mutant mice as a model to study the shared genetic etiology of neuropsychiatric diseases that may be associated with SNPs within this gene. In sum, the findings described here for the Fgf9Y162C mutant mice collectively implicate FGF9 in modulating information processing in response to a novel environment, sensorimotor recruitment, social recognition memory and adult olfactory bulb neurogenesis. In addition, this characterization expands on previous information implicating FGF9 in controlling depression- and anxiety-related behavior, highlighting its potential role in other neuropsychiatric disorders [11]. We therefore propose Fgf9Y162C mutant mice as a model to study the shared etiologies of neuropsychiatric diseases associated with this gene and as a genetic model of FGF9 dysfunction without the skeletal and respiratory aberrations that characterize existing Fgf9 genetic models. In future research, it would be propitious to characterize these mice further in terms of the underlying neural mechanisms involved in the behavioral phenotypes revealed. Acknowledgements The authors thank all the technicians from the German Mouse Clinic: Jan Einicke, Birgit Frankenberger, Sandra Geißler, Christine Hollauer, Maria Kugler, Simon Orth, Yvonne Sonntag, and Bettina Sperling as well as Erika Bürkle and Monika Stadler for the breeding of the cohorts. Thanks also to Amy Gorol for careful editing of the manuscript and to Hugh Garrett for the graphic artwork. This work has been funded by the German Federal Ministry of Education and Research to the GMC (Infrafrontier grant 01KX1012), to the German Center for Diabetes Research (DZD e.V.), the German Federal Ministry of Education and Research (BMBF) through the Integrated Network MitoPD (Mitochondrial endophenotypes of Morbus Parkinson), under the auspices of the e:Med Programme (grant 031A430E) as well as by the Helmholtz Portfolio Theme ‘Supercomputing and Modelling for the Human Brain’ (SMHB) to WW. Compliance with Ethical Standards Mice were kept under specific pathogen-free conditions at the Helmholtz Center Munich. The use of animals was in accordance with the German Law of Animal Protection,

the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and the tenets of the Declaration of Helsinki. All tests performed and described here were approved for the ethical treatment of animals by the responsible authority of the Regierung von Oberbayern (Government of Upper Bavaria).

References 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12. 13.

14.

15.

Cross-Disorder Group of the Psychiatric Genomics C, Lee SH, Ripke S, Neale BM, Faraone SV, Purcell SM, Perlis RH, Mowry BJ et al (2013) Genetic relationship between five psychiatric disorders estimated from genome-wide SNPs. Nat Genet 45(9):984– 994. doi:10.1038/ng.2711 Ford-Perriss M, Abud H, Murphy M (2001) Fibroblast growth factors in the developing central nervous system. Clin Exp Pharmacol Physiol 28(7):493–503 Mason I (2007) Initiation to end point: the multiple roles of fibroblast growth factors in neural development. Nat Rev Neurosci 8(8): 583–596. doi:10.1038/nrn2189 Terwisscha van Scheltinga AF, Bakker SC, Kahn RS, Kas MJ (2013) Fibroblast growth factors in neurodevelopment and psychopathology. Neuroscientist 19(5):479–494. doi:10.1177/ 1073858412472399 Williams AJ, Yee P, Smith MC, Murphy GG, Umemori H (2016) Deletion of fibroblast growth factor 22 (FGF22) causes a depression-like phenotype in adult mice. Behav Brain Res 307: 11–17. doi:10.1016/j.bbr.2016.03.047 Eren-Kocak E, Turner CA, Watson SJ, Akil H (2011) Short-hairpin RNA silencing of endogenous fibroblast growth factor 2 in rat hippocampus increases anxiety behavior. Biol Psychiatry 69(6): 534–540. doi:10.1016/j.biopsych.2010.11.020 Brooks LR, Enix CL, Rich SC, Magno JA, Lowry CA, Tsai PS (2014) Fibroblast growth factor deficiencies impact anxiety-like behavior and the serotonergic system. Behav Brain Res 264:74– 81. doi:10.1016/j.bbr.2014.01.053 Scearce-Levie K, Roberson ED, Gerstein H, Cholfin JA, Mandiyan VS, Shah NM, Rubenstein JL, Mucke L (2008) Abnormal social behaviors in mice lacking Fgf17. Genes Brain Behav 7(3):344–354. doi:10.1111/j.1601-183X.2007.00357.x Terauchi A, Johnson-Venkatesh EM, Toth AB, Javed D, Sutton MA, Umemori H (2010) Distinct FGFs promote differentiation of excitatory and inhibitory synapses. Nature 465(7299):783–787. doi:10.1038/nature09041 Evans SJ, Choudary PV, Neal CR, Li JZ, Vawter MP, Tomita H, Lopez JF, Thompson RC et al (2004) Dysregulation of the fibroblast growth factor system in major depression. Proc Natl Acad Sci U S A 101(43):15506–15511. doi:10.1073/pnas. 0406788101 Aurbach EL, Inui EG, Turner CA, Hagenauer MH, Prater KE, Li JZ, Absher D, Shah N et al (2015) Fibroblast growth factor 9 is a novel modulator of negative affect. Proc Natl Acad Sci U S A 112(38):11953–11958. doi:10.1073/pnas.1510456112 Lin Y, Liu G, Wang F (2006) Generation of an Fgf9 conditional null allele. Genesis 44(3):150–154. doi:10.1002/gene.20194 Colvin JS, White AC, Pratt SJ, Ornitz DM (2001) Lung hypoplasia and neonatal death in Fgf9-null mice identify this gene as an essential regulator of lung mesenchyme. Development 128(11):2095– 2106 Colvin JS, Green RP, Schmahl J, Capel B, Ornitz DM (2001) Maleto-female sex reversal in mice lacking fibroblast growth factor 9. Cell 104(6):875–889 Murakami H, Okawa A, Yoshida H, Nishikawa S, Moriya H, Koseki H (2002) Elbow knee synostosis (Eks): a new mutation

Mol Neurobiol on mouse Chromosome 14. Mamm Genome 13(7):341–344. doi: 10.1007/s00335-001-2143-6 16. Harada M, Murakami H, Okawa A, Okimoto N, Hiraoka S, Nakahara T, Akasaka R, Shiraishi Y et al (2009) FGF9 monomerdimer equilibrium regulates extracellular matrix affinity and tissue diffusion. Nat Genet 41(3):289–298. doi:10.1038/ng.316 17. Puk O, Moller G, Geerlof A, Krowiorz K, Ahmad N, Wagner S, Adamski J, de Angelis MH et al (2011) The pathologic effect of a novel neomorphic Fgf9(Y162C) allele is restricted to decreased vision and retarded lens growth. PLoS One 6(8):e23678. doi:10. 1371/journal.pone.0023678 18. Lum M, Turbic A, Mitrovic B, Turnley AM (2009) Fibroblast growth factor-9 inhibits astrocyte differentiation of adult mouse neural progenitor cells. J Neurosci Res 87(10):2201–2210. doi:10. 1002/jnr.22047 19. Fuchs H, Gailus-Durner V, Adler T, Aguilar-Pimentel JA, Becker L, Calzada-Wack J, Da Silva-Buttkus P, Neff F et al (2011) Mouse phenotyping. Methods 53(2):120–135. doi:10.1016/j.ymeth.2010. 08.006 20. Garrett L, Lie DC, de Angelis MH, Wurst W, Hölter SM (2012) Voluntary wheel running in mice increases the rate of neurogenesis without affecting anxiety-related behaviour in single tests. Bmc Neurosci 13 21. Hölter SM, Stromberg M, Kovalenko M, Garrett L, Glasl L, Lopez E, Guide J, Gotz A et al (2013) A broad phenotypic screen identifies novel phenotypes driven by a single mutant allele in Huntington’s disease CAG knock-in mice. PLoS One 8(11) 22. Stribl C, Samara A, Trumbach D, Peis R, Neumann M, Fuchs H, Gailus-Durner V, Hrabe de Angelis M et al (2014) Mitochondrial dysfunction and decrease in body weight of a transgenic knock-in mouse model for TDP-43. J Biol Chem 289(15):10769–10784 23. Zimprich A, Garrett L, Deussing JM, Wotjak CT, Fuchs H, GailusDurner V, de Angelis MH, Wurst W et al (2014) A robust and reliable non-invasive test for stress responsivity in mice. Front Behav Neurosci 8:125. doi:10.3389/fnbeh.2014.00125 24. Fuchs H, Schughart K, Wolf E, Balling R, Hrabe de Angelis M (2000) Screening for dysmorphological abnormalities—a powerful tool to isolate new mouse mutants. Mamm Genome 11(7):528–530 25. Gailus-Durner V, Fuchs H, Adler T, Aguilar Pimentel A, Becker L, Bolle I, Calzada-Wack J, Dalke C et al (2009) Systemic first-line phenotyping. Methods Mol Biol 530:463–509. doi:10.1007/978-159745-471-1_25 26. Haller F, Prehn C, Adamski J (2010) Quantification of steroids in human and mouse plasma using online solid phase extraction coupled to liquid chromatography tandem mass spectrometry. Nat Protoc. doi:10.1038/nprot.2010.1022 27. Rathkolb B, Hans W, Prehn C, Fuchs H, Gailus-Durner V, Aigner B, Adamski J, Wolf E et al (2013) Clinical chemistry and other laboratory tests on mouse plasma or serum. Curr Protoc Mouse Biol 3(2):69–100. doi:10.1002/9780470942390. mo130043 28. Garrett L, Zhang J, Zimprich A, Niedermeier KM, Fuchs H, GailusDurner V, Hrabe de Angelis M, Vogt Weisenhorn D et al (2015) Conditional reduction of adult born doublecortin-positive neurons reversibly impairs selective behaviors. Front Behav Neurosci 9: 302. doi:10.3389/fnbeh.2015.00302 29. Latchney SE, Rivera PD, Mao XW, Ferguson VL, Bateman TA, Stodieck LS, Nelson GA, Eisch AJ (2014) The effect of spaceflight on mouse olfactory bulb volume, neurogenesis, and cell death indicates the protective effect of novel environment. J Appl Physiol (1985) 116(12):1593–1604. doi:10.1152/ japplphysiol.01174.2013 30. Paxinos G, Franklin K (2001) The mouse brain in stereotaxic coordinates. Academic Press, San Diego

31.

Team RDC (2011) R: a language and environment for statistical computing. Vienna, Austria : the R Foundation for Statistical Computing http://www.R-project.org/ 32. Todo T, Kondo T, Nakamura S, Kirino T, Kurokawa T, Ikeda K (1998) Neuronal localization of fibroblast growth factor-9 immunoreactivity in human and rat brain. Brain Res 783(2):179–187 33. Couillard-Despres S, Winner B, Schaubeck S, Aigner R, Vroemen M, Weidner N, Bogdahn U, Winkler J et al (2005) Doublecortin expression levels in adult brain reflect neurogenesis. Eur J Neurosci 21(1):1–14. doi:10.1111/j.1460–9568.2004.03813.x 34. Robinson L, Plano A, Cobb S, Riedel G (2013) Long-term home cage activity scans reveal lowered exploratory behaviour in symptomatic female Rett mice. Behav Brain Res 250:148–156. doi:10. 1016/j.bbr.2013.04.041 35. Brown SD, Moore MW (2012) The International Mouse Phenotyping Consortium: past and future perspectives on mouse phenotyping. Mamm Genome 23(9–10):632–640. doi:10.1007/ s00335-012-9427-x 36. Mrosovsky N, Hampton RR (1997) Spatial responses to light in mice with severe retinal degeneration. Neurosci Lett 222(3):204– 206 37. Powell K, Ethun K, Taylor DK (2016) The effect of light level, CO2 flow rate, and anesthesia on the stress response of mice during CO2 euthanasia. Lab Anim 45(10):386–395. doi:10.1038/laban.1117 38. Swerdlow NR, Geyer MA, Braff DL (2001) Neural circuit regulation of prepulse inhibition of startle in the rat: current knowledge and future challenges. Psychopharmacology 156(2–3): 194–215 39. Rasmussen DD, Crites NJ, Burke BL (2008) Acoustic startle amplitude predicts vulnerability to develop post-traumatic stress hyper-responsivity and associated plasma corticosterone changes in rats. Psychoneuroendocrinology 33(3):282–291. doi:10.1016/j. psyneuen.2007.11.010 40. Glenn DE, Acheson DT, Geyer MA, Nievergelt CM, Baker DG, Risbrough VB, Team MRS (2016) High and low threshold for startle reactivity associated with Ptsd symptoms but not Ptsd risk: evidence from a prospective study of active duty marines. Depress Anxiety 33(3):192–202. doi:10.1002/da.22475 41. Russo AS, Parsons RG (2017) Acoustic startle response in rats predicts inter-individual variation in fear extinction. Neurobiol Learn Mem 139:157–164. doi:10.1016/j.nlm.2017.01.008 42. Pineles SL, Blumenthal TD, Curreri AJ, Nillni YI, Putnam KM, Resick PA, Rasmusson AM, Orr SP (2016) Prepulse inhibition deficits in women with PTSD. Psychophysiology 53(9):1377– 1385. doi:10.1111/psyp.12679 43. Liberzon I, Abelson JL (2016) Context processing and the neurobiology of post-traumatic stress disorder. Neuron 92(1):14–30. doi: 10.1016/j.neuron.2016.09.039 44. Powell SB, Weber M, Geyer MA (2012) Genetic models of sensorimotor gating: relevance to neuropsychiatric disorders. Curr Top Behav Neurosci 12:251–318. doi:10.1007/7854_2011_195 45. Risbrough V (2010) Behavioral correlates of anxiety. Curr Top Behav Neurosci 2:205–228 46. Lledo PM, Alonso M, Grubb MS (2006) Adult neurogenesis and functional plasticity in neuronal circuits. Nat Rev Neurosci 7:179– 193. doi:10.1038/nrn1867 47. Braun SM, Jessberger S (2014) Adult neurogenesis and its role in neuropsychiatric disease, brain repair and normal brain function. Neuropathol Appl Neurobiol 40:3–12. doi:10.1111/nan.12107 48. Bielsky IF and Young LJ (2004) Oxytocin, vasopressin, and social recognition in mammals. Peptides 25:1565–1574 49. Hoertnagl CM, Hofer A (2014) Social cognition in serious mental illness. Curr Opin Psychiatry 27:197–202 50. Woodbury ME, Ikezu T (2014) Fibroblast growth factor-2 signaling in neurogenesis and neurodegeneration. J NeuroImmune Pharmacol 9(2):92–101. doi:10.1007/s11481-013-9501-5