Creatine transporter knockout mice (Slc6a8) show increases in serotonin-related proteins and are resilient to learned helplessness

Journal Pre-proof Creatine transporter knockout mice (Slc6a8) show increases in serotonin-related proteins and are resilient to learned helplessness Zuhair I. Abdulla, Jordan L. Pennington, Arnold Gutierrez, Matthew R. Skelton

PII:

S0166-4328(19)31060-5

DOI:

https://doi.org/10.1016/j.bbr.2019.112254

Reference:

BBR 112254

To appear in:

Behavioural Brain Research

Received Date:

8 July 2019

Revised Date:

17 September 2019

Accepted Date:

18 September 2019

Please cite this article as: Abdulla ZI, Pennington JL, Gutierrez A, Skelton MR, Creatine transporter knockout mice (Slc6a8) show increases in serotonin-related proteins and are resilient to learned helplessness, Behavioural Brain Research (2019), doi: https://doi.org/10.1016/j.bbr.2019.112254

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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. © 2019 Published by Elsevier.

1

Creatine transporter knockout mice (Slc6a8) show increases in serotonin-related proteins and are resilient to learned helplessness

Zuhair I. Abdullaa,b, Jordan L. Penningtonb, Arnold Gutierreza,b, & Matthew R. Skeltona,b

aDepartment

of

of Neurology, Cincinnati Children’s Research Foundation, Cincinnati, OH USA

ro

bDivision

of Pediatrics, University of Cincinnati College of Medicine

*Correspondence:

-p

Matthew R. Skelton, Ph.D.

re

[email protected] 513-636-8632

ur na

Cincinnati, OH 45229-3039

lP

3333 Burnet Ave.

Number of pages: 24 Number of figures: 5

Number of words: Abstract 214; Introduction 439; Methods 1484; Results 464;

Jo

Discussion 1251; Figure legends 311

Author contributions: Zuhair Abdulla and Matthew Skelton designed the research, analyzed the data, and drafted the manuscript. Zuhair Abdulla, Jordan Pennington, and Arnold Gutierrez performed the research.

2

HIGHLIGHTS 

Male creatine transporter deficient mice (Slc6a8-/y) have a lack of brain Cr and alterations to the serotonergic system as evidenced by increased 5hydroxyindoleacetic acid in the striatum and hippocampus, and increased tryptophan hydroxylase and monoamine oxidase in the hippocampus.



Slc6a8-/y mice and female mice heterozygous for Slc6a8 (Slc6a8+/-) mice are resilient to learned helplessness induction. Minor alterations in anxiety-like and native-depression-like measures were

of



discovered in Slc6a8-/y and Slc6a8+/- mice.

The results of this study suggest that creatine plays a role in serotonin

ro



metabolism, which may manifest as changes in induced depressive-like

Jo

ur na

lP

re

-p

behaviors.

3

Abstract Approximately 20% of adults in the U.S. will experience an affective disorder during their life. While it is well established that serotonin (5-HT) is a crucial factor in mood, impaired cellular bioenergetics are also implicated. Creatine (Cr), through the Cr/Phospho-Cr (PCr) shuttle, maintains high ATP concentrations in the neuron. This system may be implicated in the etiology of affective disorders, as reduced Cr, PCr, and ATP are often

of

seen in the brains of affected patients. To address this issue, Cr transporter (Crt)

ro

deficient male mice (Slc6a8-/y) and female mice heterozygous for Crt expression

(Slc6a8+/-) were used to evaluate how a Cr deficient system would alter affective-like

-p

behaviors. Slc6a8-/y and Slc6a8+/- mice had faster escape latencies in learned

re

helplessness, indicating a potential resilience to behavioral despair. Slc6a8-/y had decrease latency to immobility in the tail-suspension test and Slc6a8+/- had increased

lP

open entries in elevated zero maze, but all other variables matched those of wildtype mice, however. Slc6a8-/y mice have increased 5-hydroxyindoleacetic acid content in the

ur na

hippocampus and striatum and increased monoamine oxidase protein and tryptophan hydroxylase-2 protein content in the hippocampus, while 5-HT levels are unchanged. This indicates an alteration to the 5-HTergic system in Cr deficient mice. Our results indicate that Cr plays a complex role in affective disorders and 5-HT, warranting further

Jo

investigation.

Keywords: affective disorders; depression; anxiety; energy metabolism

4

1.1 INTRODUCTION1 Affective disorders, such as depression or anxiety, are a significant worldwide healthcare burden, with depression being the leading cause of disability in the United States [1, 2]. Though primarily thought of as imbalances or deficiencies in monoaminergic activity, alterations to brain bioenergetics have also been implicated in these disorders [3, 4]. Creatine (Cr) is an important spatial and temporal ATP buffer in

of

cells with a high-energy demand [5]. Cr is transported into cells via the Cr transporter

ro

(Crt; SLC6A8) [6-11]. Supporting the role of Cr in bioenergetic pathways, Crt knockout (Slc6a8-/y) mice, which lack brain Cr have reduced brain ATP levels [10, 11]. The role of

-p

Cr on affective-type behaviors is not fully understood, though changes in brain Cr have

re

been implicated in mood disorders [12]. The Slc6a8-/y mice provide a unique opportunity to better understand the role of Cr in affective disorders, by examining the effects of a

lP

complete deficit of Cr.

Most studies related to Cr and affective behaviors have been performed following Cr

ur na

supplementation. In mice, acute Cr administration reduced immobility in the tailsuspension test (TST) [13-19]. Additionally, Cr supplemented female rats had an increased latency to immobility in the forced-swim test, while Cr-treated males were

Jo

unaffected [20]. Psychosocial defeat reduced brain Cr in tree shrews [21, 22]. In

1

Abbreviations: 5-HT: serotonin; 5-HIAA: 5-hydroxyindoleacetic acid; Cr: creatine; Crt: creatine

transporter; EZM: elevated zero maze; HAM-D: Hamilton Depression Rating Scale; LH: learned helplessness; MAO: monoamine oxidase; PCA: para-chloroamphetamine; PCr: phosphocreatine; Slc6a8-/y: male Slc6a8 knockout mice; Slc6a8+/y: male wild type mice; Slc6a8+/: female mice heterozygous for Slc6a8; Slc6a8+/+: female wild type mice; TPH-2: tryptophan hydroxylase-2; TST: tail-suspension test; cLH: rats bred for a congenital LH response

5

humans, Cr supplementation improved Hamilton Depression Rating Scale (HAM-D) scores in individuals with treatment-resistant depression [23]. Combination treatment with escitalopram and Cr increased HAM-D scores and led to a more rapid onset of symptom relief than women treated with a selective serotonin (5-HT) reuptake inhibitor (SSRI) alone [24]. Cr levels are reduced in the cerebral white matter of individuals with generalized anxiety disorder who have experienced an early life trauma [25]. Similar

of

findings were reported in the hippocampi of patients with post-traumatic stress disorder

ro

[26, 27]. Additionally, Cr supplementation increases brain Cr concentrations in both humans [28] and mice [29]. Although a link between diminished levels of Cr and

-p

affective disorders is well established, it has yet to be shown what effect a lack of Cr will

re

have on affective behaviors.

The purpose of this study was to determine if changes in affective-like behaviors

lP

occur in male Slc6a8-/y and female Slc6a8+/- mice. The female mice have heterozygous Slc6a8 expression and 50% reductions in brain Cr and mild cognitive deficits [30].

ur na

Thusly, we evaluated the effects of a complete or partial lack of Cr in the brain on behaviors related to anxiety and depression. In addition, we evaluated the effect of a loss of Cr on the expression of proteins related to 5-HT synthesis and turnover as well

Jo

as the locomotor response to a 5-HT based stimulant, para-chloroamphetamine.

2.1 METHODS

2.1.1 Subjects. Male Slc6a8-/y and female Slc6a8+/- mice, along with wild-type

littermate controls were generated in-house by breeding male Slc6a8+/y with female Slc6a8+/- mice. These mice have been maintained on their parent C57BL6/J strain with SNP analysis showing 97% similarity to the stock C57BL6/J line at Jackson Laboratory.

6

Homozygous knockout female mice cannot be generated since Slc6a8-/y mice do not breed. Mice were genotyped by PCR as previously described [10, 11]. Mice were housed at 21° C in microisolators 2-4/cage with ad libitum access to food and water. Lights were maintained on a 14 h:10 h light:dark cycle with lights on at 0700. All procedures on mice were performed at the Cincinnati Children’s Research Foundation vivarium which is fully accredited by AAALAC International and protocols were approved

of

by the Institutional Animal Care and Use Committee, protocol #2017-0019. All

ro

institutional and national guidelines for the care and use of laboratory animals were

followed. No randomization was performed to allocate subjects in the study. Adult mice

-p

aged 60 to 80 days were assigned by litter to one of three behavioral cohorts, with only

re

one animal per genotype per litter used. The first cohort was tested in elevated zero maze (EZM) and tail-suspension test (TST) in that order. Cohort 2 were assessed for

lP

learned helplessness (LH), while a third cohort was used for the PCA challenge. Only tissue from cohort 1 was used for protein analyses. Data were collected from a total of

ur na

199 mice, with no sample calculations having been performed; instead sample sizes were determined by previously collected data. A flow chart showing how animals were used is included as supplemental information. No exclusion criteria were predefined. All behavioral testing was run between 1000 and 1400 o’clock on mice.

Jo

2.1.2 Elevated Zero Maze. The protocol followed those previously published with

minor modifications [31-33]. The maze consists of a circular runway [61 cm (dia) × 5.1 cm (width), 48.5 cm above the ground] with two walled corridors (inner wall [16 cm (h) × 42 cm (l)] and outer walls [16 cm (h) × 49 cm (l)]) on opposite sides, creating alternating open and closed quadrants. The room lights were dimmed to 26.5 lux. Mice were placed in a closed quadrant and the experimenter left the room. Using a closed-circuit digital

7

camera system, behavior was scored in the hallway outside of the procedure room. Latency to the first open quadrant entrance, total time spent in open quadrants, number of distinct open quadrant entries, and number of head dips were live scored by an experimenter blinded to the treatment using ODLog software (Macropod Software). 2.1.3 Tail-Suspension Test. The tail-suspension apparatus consists of a transparent platform [33 cm (l) × 20 cm (d) x 16.5 cm (h)] with 4 holes (0.635 cm diameter) covered

of

by rubber grommets [32]. At the beginning of the 5 min (300 s) trial, the mouse’s tail was

ro

passed through a hole from the underside of the platform and the mouse was raised to the point where it was unable to reach up and grasp the base of its tail for support. The

-p

tail was taped to the top of the platform. Trials were live-scored for latency to first

re

immobility, total mobility time, and total immobility time [34].

2.1.4 PCA Challenge. Baseline and stimulant-induced locomotor activity were

lP

monitored using the PAS system from San Diego Instruments (San Diego, CA). The 41 × 41-cm chambers are made of clear acrylic and are lined with horizontal strips of 16

ur na

LED infrared beams, with each beam spaced 2.56 cm apart. Mice were placed in the chambers for 30 min to measure baseline activity. Following baseline, mice received an i.p. injection of saline (10 ml/kg) and were placed back into the chamber for 30 min to control for injection stress. Mice then received 1 mg/kg, 2.5 mg/kg, or 5 mg/kg of the 5-

Jo

HT agonist para-chloroamphetamine (PCA; volume 10 mg/ml, adjusted for free base) and were placed back into the arena for 2 h. Unique and repetitive beam breaks were recorded by the PAS software. Total ambulatory activity is calculated by taking the sum of all unique beam breaks. Data are analyzed as the sum of total activity within 5 min blocks.

8

2.1.5 Learned Helplessness. Our protocol was adapted from Hufgard et al. 2017 [35]. LH was tested using the GEMINI Avoidance System (San Diego Instruments, San Diego, CA). The apparatus consists of an arena with two chambers [size of each chamber: 24 cm (w) × 20.5 cm (d) × 20.5 cm (h)] separated by a mechanical gate. To ensure that the paradigm induced LH, mice were assigned to one of two treatment groups: pre-stressed or control. On the two training days, mice were placed in the

of

chamber with the gate closed for approximately 45 min. Mice in the SHOCK group

ro

received 360 foot-shocks (0.3 mA, 2 s duration, randomized intertrial intervals (ITI) of 1– 10 s) that were omitted in the control group. On day three, both groups were tested for

-p

learned helplessness. Mice were placed in the original chamber to start the trial; the

re

gate was lifted and a continuous foot shock (0.3 mA) was delivered until the mouse escaped to the adjacent chamber or until the time limit (24 s) had elapsed. Mice

were recorded.

lP

received 30 trials with a 10 s ITI. Latency to escape and trials with a successful escape

ur na

2.1.6 High Performance Liquid Chromatography. Mice tested in the EZM and TST were used for analysis of neurotransmitter content. Within 1 week of testing, mice were anesthetized with isoflurane, decapitated, and the brain was rapidly removed. The same experimenter performed all dissections to control for the time from dissection to freezing.

Jo

The hippocampus and striatum tissue were dissected free-hand, flash frozen in dry-ice cooled methylbutane, and stored at -80° C until processing. Tissue was sonicated in icecold 0.1 N perchloric acid (300 µl per striatum, 500 µl per hippocampus) and centrifuged at 20,800 RCF for 13 min at 4.0o C. Monoamine concentrations were measured from the supernatant using high-performance liquid chromatography (HPLC). The HPLC-ECD system consists of a Dionex UltiMate® 3000 Analytical Autosampler connected to an

9

ESA 5840 pump and a Coulochem III electrochemical detector (Thermo Scientific), which was set to -150 mV for E1 and +250 mV for E2. The guard cell was set to +350 mV. The column used was a Supelco Supelcosil™ LC-18 column (15 cm x 4.6 mm, 3 μm; Sigma-Aldrich Co.). MD-TM Mobile Phase (Thermo Fisher Scientific; 89% water, 10% acetonitrile, and 1% sodium phosphate monobasic (monohydrate)) was used. The flow rate was 0.5 ml/min. Serial dilutions of standards from each monoamine were used

of

to generate a standard curve and determine the concentration of the tissue samples.

ro

2.1.7 Western Blots. Hippocampi from all behavioral groups were homogenized by sonication in RIPA buffer with protease inhibitors (composed of sodium chloride, sodium

-p

dodecyl sulfate, sodium deoxycholate, Tris-HCL, Triton X). Protein concentrations were

re

determined using the Pierce bicinchoninic acid assay (BCA; Thermo-Fisher). Samples were mixed with 2x Laemmli buffer (Life Sciences) to a final concentration of 1 μg/μL

lP

and incubated at 95° C for 5 min. A total of 10 µg protein/well was separated by electrophoresis on 12% Mini-PROTEAN TGX gels (Bio-Rad). Proteins were transferred

ur na

to Immobilon-FL PVDF membranes (Millipore) followed by incubation in Odyssey blocking buffer (LI-COR, Lincoln, NE) for 1 h at room temperature (RT). Blots were incubated with antibodies for TPH-2 (Abcam, ab184505), monoamine oxidase A (MAO; Abcam, ab126751), α-tubulin (Sigma Aldrich, T5168), and ß-actin (LI-COR, #926-42212)

Jo

overnight with agitation at 4° C. Membranes were washed with TBST and incubated at RT with IRDye (LI-COR) secondary antibodies for 1 h. Membranes were again washed with TBS-T followed by TBS and visualized using an Odyssey CLx scanner (LI-COR). Bands were analyzed in Image Studio (LI-COR). TPH-2 and MAO band intensities were normalized to ß-actin.

10

2.1.8 Statistics. Male and female mice have distinct genotypes, so using sex as a between-subjects factor would create an imbalanced design as there are no heterozygous males and no homozygous females. Therefore, each sex is analyzed separately with genotype as the only between subjects factor. Statistical analysis was performed using GraphPad Prism (San Diego, CA) software. For experiments with a single variable and where the two groups were compared, unpaired Student’s t-tests

of

were used. For tests involving measurement of multiple time blocks, e.g. locomotor

ro

activity and learned helplessness, a repeated-measures analysis of variance (ANOVA) was used with gene as the between subjects factor and time as the repeated factor.

-p

Significant effects were analyzed using the False Discovery Rate method of Benjamini,

re

Krieger, and Yekutieli [36]. Appropriate nonparametric analyses were used when data did not fit a Gaussian distribution and are noted in the text. The null hypothesis that the

lP

groups are the same was rejected at p<0.05 for all tests. Outliers were defined as values lying more than three standard deviations from the mean in either direction. Values

ur na

defined as outliers were removed from analysis.

3.1 RESULTS

3.1.1 Weights. Slc6a8-/y mice weighed less than Slc6a8+/y mice at the start of testing

Jo

((MEAN±SEM: Slc6a8+/y= 24.8±0.3 g; Slc6a8-/y=15.5±0.4 g); t36=19.12. p<0.0001). No differences were observed between the female Slc6a8+/+ (MEAN±SEM=20.1±0.35 g) and Slc6a8+/- (MEAN±SEM=19.19±0.35 g) mice. 3.1.2 Elevated Zero Maze. In males, there was no effect of gene for any variable examined. Female Slc6a8+/- mice had more entries into the open arms than Slc6a8+/+ mice (t38=2.153, p<0.05; Figure 1c). No differences were observed for other variables.

11

Four outliers in Slc6a8+/y and 3 each in Slc6a8-/y, Slc6a8+/+, and Slc6a8+/- were discovered in the latency to open variable and were removed from all EZM analysis. 3.1.3 Tail-Suspension Test. Slc6a8-/y had shorter latencies to first immobility than Slc6a8+/y mice (t36=2.12, p<0.05; Figure 2a). There was a time × gene interaction for mobility (F(3,144)=4.58, p<0.01; Figure 2b), but post hoc analyses did not indicate a difference at any time point. A significant time × gene interaction for immobility times

of

was discovered for male mice (F(3, 144)=3.52, p<0.01; Figure 2c), however, post hoc

ro

analyses yielded no subsequent differences. There were no differences amongst female mice. One female outlier from each group was identified in latency to first immobility and

-p

removed from all TST analyses.

re

3.1.4 Learned Helplessness. For males, the three-way interaction of gene x shock x trial was significant (F(29, 986)=2.063, p<0.001). The three-way interaction was

lP

significant in females as well (F(29, 1247)=2.204, p<0.001). To determine if the paradigm induced learned helplessness in each genetic group, shock and trial were

ur na

examined by genotype. In Slc6a8+/y mice, there was a main effect of shock (F(1,20)=55.92, p<0.0001) and a shock x trial interaction (F(29,580)=2.105 p<0.001) with mice in the shock group having longer latencies for each trial compared with mice in the no shock group (p<0.05 for each trial; Figure 3a). Slc6a8-/y mice in the shock group

Jo

had longer overall latencies to escape compared with mice from the no shock group (F(1,14)=7.550, p<0.05); however, while there was shock x trial interaction (F(29,406)=1.838, p<0.001), none of the individual trials were significantly different (Figure 3b). Longer escape latencies were observed for Slc6a8+/+ mice in the shock group compared with no shock mice of the same genotype (F(1,26)=11.82, p<0.01; Figure 3c). When examining each trial (shock x time interaction (F(29,754)=1.703,

12

p<0.05), the Slc6a8+/+-shock mice consistently had faster escape latencies than their non-shocked counterparts (all trials 1-30 p<0.05). There were no significant main effects or an interaction of shock and trial in the Slc6a8+/- mice (Figure 3d). Within the shock groups, the Slc6a8-/y mice had shorter escape latencies than Slc6a8+/y mice (F(1,22)=6.149, p<0.05; Figure 3e). A gene x time interaction (F(29,638)=2.554, p<0.0001) showed that Slc6a8-/y mice were faster to escape than Slc6a8+/y mice on trials

of

1-5 and 7 (p<0.05). In the female shock group, Slc6a8+/- mice had faster escape

ro

latencies than Slc6a8+/+ mice (F(1,29)=5.323, p<0.05; Figure 3f). No by gene differences

-p

were observed in the no shock group for male or female mice.

Slc6a8-/y mice had more trials with an escape than Slc6a8+/y mice (gene x shock

re

interaction: F(1,43)=7.03, p<0.05). A similar interaction (gene x shock interaction: F(1,51)=4.07, p<0.05) was seen in the number of trials with an escape with Slc6a8+/-

lP

mice escaping more than Slc6a8+/+ mice

3.1.5 PCA Challenge. The was no change in activity levels following administration of

ur na

1.0 or 2.5 mg/kg of PCA (Figure 4a, b). PCA administration at 5 mg/kg produced hyperactivity in both groups (main effect of drug, saline vs PCA; (F(1,48)=50.58, p<0.0001); Figure 4c), however there were no differences between Slc6a8-/y and

Jo

Slc6a8+/y mice (Figure 4c).

3.1.6 Monoamine assessment. Males: Hippocampal 5-HIAA levels were increased in

Slc6a8-/y mice compared with Slc6a8+/y mice (t22=2.384, p<0.05; Figure 4b). A similar increase was seen in the neostriatum for 5-HIAA (t21=2.488, p<0.01; figure 4e). In addition, increases in striatal DA (t21=2.493, p<0.05; Figure 4g) and 3-HVA (t21=2.246, p<0.0356; Figure 4i) were observed in Slc6a8-/y mice compared with Slc6a8+/y mice. All

13

other neurotransmitters evaluated were unchanged. Females: No differences were found in either brain region for any neurotransmitter or their metabolites. 3.1.7 5-HT markers. Males: TPH-2 signal intensity was increased in the hippocampus of Slc6a8-/y mice compared with Slc6a8+/y mice (t21=1.872, p<0.05, Figure 5a). Hippocampal MAO signal intensity was increased in Slc6a8-/y vs Slc6a8+/y mice

ro

differences were observed in TPH-2 or MAO expression levels.

of

(t23=1.93, p<0.05; Figure 5b). Tissue from the striatum was not analyzed. Females: No

4.1 DISCUSSION

-p

The purpose of this study was to evaluate affective-like behaviors in Crt-deficient

re

mice. Slc6a8-/y mice express limited LH, an induced depressive-like behavior, while Slc6a8+/- appear to be completely resistant to LH-induction. Additionally, male Slc6a8-/y

lP

mice have increased 5-HIAA and increased expression of TPH-2 and MAO, proteins related to 5-HT synthesis and metabolism. Performance in anxiety-like or native

ur na

depressive-like behavior, as evaluated by the EZM and TST respectively, were not drastically altered by Crt-deficiency: Slc6a8-/y mice had shorter latencies to immobility in the TST and Slc6a8+/- mice had increased open quadrant entries in EZM. Together, the results of this study suggest that Cr plays a role in 5-HT metabolism, which may

Jo

manifest as changes in induced depressive-like behaviors. The LH task represents an animal model of coping strategies that has a high

degree of translational value related to depression as it is seen in humans and rodent models [37]. Accordingly, treatment with antidepressants leads to resilience against LH, providing further face validity to this model [38]. The 5-HT system has been shown to play an important role in the acquisition of LH [37]. Additionally, Lugenbiel et al. (2010)

14

found that rats bred for a congenital LH response (cLH) without induction have decreased Slc6a8 expression in the hippocampus, but not the prefrontal cortex [39]. Both escitalopram and electroconvulsive shock administration are antidepressant treatments that attenuate LH-responses [40, 41]. Interestingly, Lugenbiel at al. demonstrated that these treatments increased Slc6a8 expression in both the hippocampus and prefrontal cortex of cLH, but not wild type, rats. Conversely, this study

of

found that mice with either a lack of Slc6a8 expression were more resilient against LH-

ro

induction than wild type mice, and that female mice with reduced Slc6a8 expression are resistant to LH-induction. While the results of this study do not match the results of the

-p

Lugenbiel study, there are caveats that could prevent a direct comparison of these

re

studies. While Slc6a8 expression was reduced in the hippocampus, Cr levels were not measured in the cLH rats. In addition, the expression of Slc6a8 was unchanged in the

lP

PFC of cLH rats [39]. On the other hand, Slc6a8-/y have a total lack of brain Crt and Cr [11], while female Slc6a8+/- mice have a mosaic expression of Crt and 50% reductions of

ur na

brain Cr [30]. It is possible that the differences in these studies could be a result of changes caused by a lack of Cr in the PFC. Finally, Slc6a8 expression was not eliminated in the rats tested by Lugenbiel et al. (2010), while it is absent in the Slc6a8-/y mice. It is possible that reductions in Slc6a8 and elimination of the gene lead to distinct

Jo

phenotypes in male rodents. Female response to LH and subsequent Crt expression was not assessed by Lugenbiel et al. (2010) [39]. Given the significant cognitive deficits in Slc6a8-/y mice [11] and the moderate deficits of Slc6a8+/- mice [30], an alternative to the resiliency explanation is that these mice lack the cognitive ability to learn that the shocks were unescapable. It should be noted, that the female Slc6a8+/- mice were the most resistant to the LH protocol however do not show deficits in contextual or

15

conditioned fear memory which is induced by footshock exposure [30]. Contextual and conditioned fear deficits are present in Slc6a8-/y mice [11]. This would suggest that the female mice are averse to and retain the memory of a footshock and that the differences between the groups are not likely due to memory deficits. A protocol with more training days prior to the LH assessment could indicate whether this effect is cognitively or affectively governed. In addition, the use of cell-specific manipulations, like a 5-HT

of

specific Slc6a8 knockout, could allow for the dissociation of the learning deficits from the

ro

depressive-like aspects of LH testing.

It is possible that Cr deficiency has a mild effect on depressive- or anxiety-like

-p

behavior. The TST is used to assess depressive-like behaviors, interpreted as increased

re

immobility during the task. Mobility is thought to represent escape behavior, or struggling, while immobility is thought of as a form of behavioral despair [42]. Slc6a8-/y

lP

mice had a decreased latency to immobility in the TST, but overall immobility was unaffected. The similar immobility times indicate that Slc6a8-/y and Slc6a8+/y mice

ur na

exhibited equal amounts of behavioral despair, but only displayed it earlier. It is possible then that this was a motor effect, as Slc6a8-/y mice are significantly smaller than wild type mice [11]. EZM is a tool used to measure anxiety-like behaviors, as mice are instinctually reluctant to enter open spaces and to prefer the safety of the high walls of

Jo

the closed quadrants [43]. Slc6a8+/- mice had more open quadrant entries in EZM than Slc6a8+/+ mice, although all other variables of EZM were unchanged. Even so, open quadrant entries are considered a fairly sensitive measure of anxiety-like resistance [43]. The lack of change in the other variables of these tasks, and the sex specificity of the effects, prevents us from ascribing a causative effect of Cr deficiency on native depression or anxiety like behaviors.

16

The minor changes observed in the TST and EZM seem to be incompatible with the resilience to LH-induction in Slc6a8-/y and Slc6a8+/- mice. However, these seemingly divergent results are not altogether surprising as these tests measure different affective outcomes. A crucial difference between LH and our other affective measurements is that TST and EZM were measured in mice that were not exposed to a known stressor prior to testing. This was done to ensure that any potential alterations in behavior would

of

provide insight into the baseline affective behaviors in mice lacking Slc6a8. In wild type

ro

mice, a depressive-like phenotype is often induced by exposure to known stressors prior to testing [44]. Based on the differences in the tasks used in this study, it may be

-p

beneficial to employ these strategies in Slc6a8-/y or Slc6a8+/- mice in future studies.

re

We have shown that Slc6a8-/y and Slc6a8+/- mice have cognitive deficits that are caused by the lack of brain Cr [11, 45]. It is important to determine what other behaviors

lP

may be disrupted by the lack of Cr. In the initial characterization of this model, social interaction and acoustic startle/pre-pulse inhibition were not changed in Slc6a8-/y mice

ur na

[11]. The results from the TST and EZM further establish that the loss of Cr does not lead to a global behavioral deficit. These findings provide guidance for a more precise determination of pathways and brain regions that could be disrupted in the Cr-deficient brain.

Jo

It is well-established that LH is dependent on 5-HT function. In the mice used for

the initial characterization in Skelton et al. (2011), there was a small (11%) increase in hippocampal 5-HT levels of Slc6a8-/y mice along with a more pronounced (34%) increase in 5-HIAA. In this study, while there was a small increase (5%) in the mean 5HT content of the hippocampus, it did not reach significance. Given the magnitude of the change between the studies and the larger sample size of this study, it would not be

17

surprising that a type-1 error was present in the 2011 study. The increase in 5-HIAA was replicated between the studies with Slc6a8-/y mice having a 25% increase in hippocampal 5-HIAA levels. Additionally, we observed an increase in MAO protein levels in the hippocampus of Slc6a8-/y mice. MAO is an enzyme which metabolizes 5-HT [46]. Hippocampal TPH-2 protein was also increased in Slc6a8-/y mice. TPH-2 is an enzyme involved in the biosynthesis of 5-HT in the brain [47, 48]. The limited or lack of change in

of

5-HT along with the increase in 5-HIAA, when coupled with the elevated TPH-2 and

ro

MAO could suggest that 5-HT turnover is increased in Slc6a8-/y mice. However, the ratio of 5-HT:5-HIAA and the protein changes do not directly show that 5-HT release or

-p

metabolism is altered, though alternative interpretations of these findings would be

re

unlikely. Using methods that can measure 5-HT release, such as microdialysis or fastscan cyclic voltammetry, in the future could provide valuable information on the role of

lP

Cr in neurotransmitter release. One possible explanation as to how these 5-HTergic alterations occurred is that they were caused by changes in brain energy metabolism. Cr

ur na

acts as a buffer to maintain ATP levels in cells with high energy demand, including neurons [5]. As such, Slc6a8-/y mice have decreased brain ATP concentrations relative to Slc6a8+/y mice. Given that most of the brain’s energy is devoted to synaptic transmission, it is likely that the energetic alterations present within Slc6a8-/y mice led to

Jo

the 5-HT alterations. Although they have a 50% reduction in brain Cr, similar serotonergic alterations were not observed in female Slc6a8+/- mice. Because both Slc6a8+/- and Slc6a8-/y mice displayed a resilience to LH, yet only Slc6a8-/y mice have altered 5-HT metabolism this study is unable to determine what role, if any, 5-HT metabolism has in LH-induction resiliency. Due to the lack of differences in 5-HT metabolism, female mice were not included in the PCA challenge.

18

Acute doses of PCA (0.5 to 5 mg/kg) increase synaptic 5-HT and induce hyperactivity in mice [49, 50]. Interestingly, there were no significant alterations to motor activity in Slc6a8-/y mice after PCA administration. It is possible that the high dose of 5 mg/kg PCA caused a ceiling effect with maximal 5-HT release. However, dosing at 1 mg/kg and 2.5 mg/kg failed to produce hyperactivity in either group. Alternatively, the higher dose could have caused an interaction with other neurotransmitter pathways.

of

While PCA has a relatively high affinity for the 5-HT system, it also increases dopamine

ro

release at higher doses [51]. Therefore, PCA co-administered with DA receptor

antagonists could provide valuable data to better understand these interactions.

-p

Additionally, use of a more specific 5-HT agonist, such as 8-OH-DPAT, a compound that

re

favors the 5-HT1a receptors that are important for mood and depression-like behaviors [18], could provide a more nuanced finding.

lP

The main findings of this study were that Cr deficiency leads to resiliency in LH and that a total loss of Cr causes augmentations to the serotonergic system without

ur na

creating an overt endogenous depression- or anxiety-like phenotype. Given that Cr supplementation in both humans and rodents can reduce depressive symptoms, the lack of an overtly depressive phenotype in Cr deficient mice is unexpected. However, the results of this study do not invalidate the wealth of evidence suggesting that Cr

Jo

supplementation is quite often successfully used as an adjunct to antidepressant medication. These mice show significant cognitive deficits [11, 45], in agreement with humans with X-linked Crt deficiency (CTD) who have moderate to severe intellectual disability and delayed language development [52-55]. While the cognitive phenotype of human CTD patients and Slc6a8-/y mice are well-established, the results of this study provide a greater understanding of CTD by refining the behavioral profile of this

19

important mouse model. Interestingly, humans with CTD are often described as having a “happy nature” and affective disorders have not been identified as a co-morbidity of CTD [54]. Therefore, results of this study further solidify this mouse as a high-fidelity model of CTD.

Funding: This work was supported by NIH grant HD080910, a CARE grant from the

of

Association for Creatine Deficiencies, and internal support from the Division of

ro

Neurology (MRS).

-p

Acknowledgements: The authors thank Marla K. Perna for editorial support. The

Jo

ur na

lP

re

authors declare no conflicts of interests.

20

References 1.

NIMH, Top 10 Leading Disease/Disorder Categories Contributing to U.S. DALYs. 2010.

2.

NIMH, U.S. DALYs for Mental and Behavioral Disorders as a Percent of Total U.S. DALYs. 2010. Gardner, A. and R.G. Boles, Beyond the serotonin hypothesis: mitochondria,

of

3.

inflammation and neurodegeneration in major depression and affective spectrum

Klinedinst, N.J. and W.T. Regenold, A mitochondrial bioenergetic basis of

-p

4.

ro

disorders. Prog Neuropsychopharmacol Biol Psychiatry, 2011. 35(3): p. 730-43.

depression. J Bioenerg Biomembr, 2015. 47(1-2): p. 155-71. Wyss, M. and R. Kaddurah-Daouk, Creatine and Creatinine Metabolism. Physiol Rev, 2000. 3(80): p. 1107-1213.

Braissant, O., et al., Creatine deficiency syndromes and the importance of

lP

6.

re

5.

creatine synthesis in the brain. Amino Acids, 2011(40): p. 1315-1324. Braissant, O., et al., Endogenous synthesis and transport of creatine in the rat

ur na

7.

brain: an in situ hybridization study. Molecular Brain Research, 2001. 86(1-2): p. 193-201. 8.

Hanna-El-Daher, L. and O. Braissant, Creatine synthesis and exchanges

Jo

between brain cells: What can be learned from human creatine deficiencies and various experimental models? Amino Acids, 2016. 48(8): p. 1877-95.

9.

Joncquel-Chevalier Curt, M., et al., Creatine biosynthesis and transport in health and disease. Biochimie, 2015. 119: p. 146-65.

21

10.

Perna, M.K., et al., Creatine transporter deficiency leads to increased whole body and cellular metabolism. Amino Acids, 2016. 48(8): p. 2057-65.

11.

Skelton, M.R., et al., Creatine transporter (CrT; Slc6a8) knockout mice as a model of human CrT deficiency. PLoS One, 2011. 6(1): p. e16187.

12.

Allen, P.J., Creatine metabolism and psychiatric disorders: Does creatine supplementation have therapeutic value? Neurosci Biobehav Rev, 2012. 36(5): p.

Cunha, M.P., et al., Involvement of PI3K/Akt Signaling Pathway and Its

ro

13.

of

1442-62.

Mol Neurobiol, 2016. 53(5): p. 2954-2968.

Cunha, M.P., et al., Involvement of PKA, PKC, CAMK-II and MEK1/2 in the acute

re

14.

-p

Downstream Intracellular Targets in the Antidepressant-Like Effect of Creatine.

antidepressant-like effect of creatine in mice. Pharmacol Rep, 2014. 66(4): p.

15.

lP

653-9.

Cunha, M.P., et al., Antidepressant-like effect of creatine in mice involves

16.

ur na

dopaminergic activation. J Psychopharmacol, 2012. 26(11): p. 1489-501. Cunha, M.P., et al., The modulation of NMDA receptors and L-arginine/nitric oxide pathway is implicated in the anti-immobility effect of creatine in the tail suspension test. Amino Acids, 2015. 47(4): p. 795-811. Cunha, M.P., et al., The activation of alpha1-adrenoceptors is implicated in the

Jo

17.

antidepressant-like effect of creatine in the tail suspension test. Prog Neuropsychopharmacol Biol Psychiatry, 2013. 44: p. 39-50.

18.

Cunha, M.P., et al., Evidence for the involvement of 5-HT1A receptor in the acute antidepressant-like effect of creatine in mice. Brain Res Bull, 2013. 95: p. 61-9.

22

19.

Pazini, F.L., et al., Creatine, Similar to Ketamine, Counteracts Depressive-Like Behavior Induced by Corticosterone via PI3K/Akt/mTOR Pathway. Mol Neurobiol, 2016. 53(10): p. 6818-6834.

20.

Allen, P.J., et al., Sex-specific antidepressant effects of dietary creatine with and without sub-acute fluoxetine in rats. Pharmacol Biochem Behav, 2012. 101(4): p. 588-601. Czeh, B., et al., Stress-induced changes in cerebral metabolites, hippocampal

of

21.

ro

volume, and cell proliferation are prevented by antidepressant treatment with tianeptine. Proc Natl Acad Sci, 2001. 98(22): p. 12796-12801.

van der Hart, M.G., et al., Substance P receptor antagonist and clomipramine

-p

22.

re

prevent stress-induced alterations in cerebral metabolites, cytogenesis in the dentate gyrus and hippocampal volume. Mol Psychiatry, 2002. 7(9): p. 933-41. Roitman, S., et al., Creatine monohydrate in resistant depression: a preliminary

lP

23.

study. Bipolar Disorders, 2007(9): p. 754-758. Lyoo, I.K., et al., A Randomized, Double-Blind Placebo-Controlled Trial of Oral

ur na

24.

Creatine Monohydrate Augmentation for Enhanced Response to a Selective Serotonin Reuptake Inhibitor in Women With Major Depressive Disorder. Am J Pyschiatry, 2012(169): p. 937-945. Coplan, J.D., et al., Decreased choline and creatine concentrations in centrum

Jo

25.

semiovale in patients with generalized anxiety disorder: relationship to IQ and early trauma. Psychiatry Res, 2006. 147(1): p. 27-39.

26.

Villarreal, G., et al., Proton Magnetic Resonance Spectroscopy of the Hippocampus and Occipital White Matter in PTSD: Preliminary Results. Can J Psychiatry, 2002. 47(7): p. 666-670.

23

27.

Schuff, N., et al., Abnormal N-acetylaspartate in hippocampus and anterior cingulate in posttraumatic stress disorder. Psychiatry Res, 2008. 162(2): p. 14757.

28.

Dechent, P., et al., Increase of total creatine in human brain after oral supplementation of creatine-monohydrate. American Journal of PhysiologyRegulatory, Integrative and Comparative Physiology, 1999. 277(3): p. R698-

Ipsiroglu, O.S., et al., Changes of tissue creatine concentrations upon oral

ro

29.

of

R704.

Sciences, 2001. 69(15): p. 1805-1815.

Hautman, E.R., et al., Female mice heterozygous for creatine transporter

re

30.

-p

supplementation of creatine-monohydrate in various animal species. Life

deficiency show moderate cognitive deficits. J Inherit Metab Dis, 2014. 37(1): p.

31.

lP

63-8.

Amos-Kroohs, R.M., et al., Neurobehavioral phenotype of C57BL/6J mice

ur na

prenatally and neonatally exposed to cigarette smoke. Neurotoxicol Teratol, 2013. 35: p. 34-45. 32.

Schaefer, T.L., C.V. Vorhees, and M.T. Williams, Mouse plasmacytomaexpressed transcript 1 knock out induced 5-HT disruption results in a lack of

Jo

cognitive deficits and an anxiety phenotype complicated by hypoactivity and defensiveness. Neuroscience, 2009. 164(4): p. 1431-43.

33.

Sprowles, J.L., et al., Perinatal exposure to the selective serotonin reuptake inhibitor citalopram alters spatial learning and memory, anxiety, depression, and startle in Sprague-Dawley rats. Int J Dev Neurosci, 2016. 54: p. 39-52.

24

34.

Can, A., et al., The Tail Suspension Test. Journal of Visualized Experiments, 2012(59).

35.

Hufgard, J.R., M.T. Williams, and C.V. Vorhees, Phosphodiesterase-1b deletion confers depression-like behavioral resistance separate from stress-related effects in mice. Genes, Brain and Behavior, 2017.

36.

Benjamini, Y., A.M. Krieger, and D. Yekutieli, Adaptive linear step-up procedures

Maier, S.F. and M.E. Seligman, Learned helplessness at fifty: Insights from

ro

37.

of

that control the false discovery rate. Biometrika, 2006. 93(3): p. 491-507.

neuroscience. Psychol Rev, 2016. 123(4): p. 349-67.

Chourbaji, S., et al., Learned helplessness: validity and reliability of depressive-

-p

38.

39.

re

like states in mice. Brain Res Brain Res Protoc, 2005. 16(1-3): p. 70-8. Lugenbiel, P., et al., Creatine transporter expression after antidepressant therapy

lP

in rats bred for learned helplessness. The World Journal of Biological Psychiatry, 2010. 11(2): p. 329-333.

Reed, A.L., et al., Treatment with escitalopram but not desipramine decreases

ur na

40.

escape latency times in a learned helplessness model using juvenile rats. Psychopharmacology (Berl), 2009. 205(2): p. 249-59. 41.

Sartorius, A., et al., Specific creatine rise in learned helplessness induced by

Jo

electroconvulsive shock treatment. Neuroreport, 2003. 14(17): p. 2199-201.

42.

Steru, L., et al., The tail suspension test: A new method for screening antidepressants in mice. Psychopharmacology, 1985. 85: p. 367-370.

43.

Shepherd, J.K., et al., Behavioural and pharmacological characterisation of the elevated "zero-maze" as an animal model of anxiety. Psychopharmacology, 1994(116): p. 56-64.

25

44.

Hammels, C., et al., Defeat stress in rodents: From behavior to molecules. Neurosci Biobehav Rev, 2015. 59: p. 111-40.

45.

Udobi, K.C., et al., Cognitive deficits and increases in creatine precursors in a brain-specific knockout of the creatine transporter gene Slc6a8. Genes Brain Behav, 2018. 17(6): p. e12461.

46.

Jonnakuty, C. and C. Gragnoli, What do we know about serotonin? J Cell Physiol,

Sakowski, S.A., et al., Differential tissue distribution of tryptophan hydroxylase

ro

47.

of

2008. 217(2): p. 301-6.

isoforms 1 and 2 as revealed with monospecific antibodies. Brain Res, 2006.

Gaspar, P. and C. Lillesaar, Probing the diversity of serotonin neurons. Philos

re

48.

-p

1085(1): p. 11-8.

Trans R Soc Lond B Biol Sci, 2012. 367(1601): p. 2382-94. Fuller, R.W., Effects of p-Chloroamphetamine on Brain Serotonin Neurons.

lP

49.

Neurochemical Research, 1992. 17(5): p. 449-456. Geyer, M.A., Serotonergic functions in arousal and motor activity. Behavioural

ur na

50.

Brain Research, 1996(73): p. 31-35. 51.

Henderson, M.G., K.W. Perry, and R.W. Fuller, Possible Involvement of Dopamine in the Long Term Serotonin Depletion by p-Chloroamphetamine and

Jo

beta,beta-Difluoro p-Chloroamphetamine in Rats. The Journal of Pharmacology and Experimental Therapeutics, 1993. 267(1): p. 417-424.

52.

van de Kamp, J.M., et al., Phenotype and genotype in 101 males with X-linked creatine transporter deficiency. J Med Genet, 2013. 50(7): p. 463-72.

26

53.

van de Kamp, J.M., et al., New insights into creatine transporter deficiency: the importance of recycling creatine in the brain. J Inherit Metab Dis, 2013. 36(1): p. 155-6.

54.

van de Kamp, J.M., G.M. Mancini, and G.S. Salomons, X-linked creatine transporter deficiency: clinical aspects and pathophysiology. J Inherit Metab Dis, 2014. 37(5): p. 715-33. van de Kamp, J.M., et al., Long-term follow-up and treatment in nine boys with X-

of

55.

Jo

ur na

lP

re

-p

ro

linked creatine transporter defect. J Inherit Metab Dis, 2012. 35(1): p. 141-9.

27

Figure Legends

Figure 1. Elevated zero maze. (a) Latency to open quadrant entry, (b) time in open quadrants, and (d) head dips were unchanged. (c) Slc6a8+/- mice entered the open quadrants more often than Slc6a8+/+ mice (t38=2.153, p<0.05). There were 4 outliers in Slc6a8+/y and 3 each in Slc6a8-/y, Slc6a8+/+, and Slc6a8+/-. n=17-20/group, * p<0.05

of

Figure 2. Tail-suspension test. (a) Slc6a8-/y mice had a decreased latency to first

ro

immobility compared with Slc6a8+/y mice (t36=2.12, p<0.05). In males (b and c), total time mobile (b) and immobile (c) were similar between groups. Similar effects were seen in

-p

females (d and e, respectively). One outlier was removed from each of the female

re

groups. n=20-22/group, * p<0.05

Figure 3. Learned Helplessness. Significant differences between SHOCK and control

lP

groups showed that our protocol resulted in successful LH-induction in both male groups (a) and Slc6a8+/+ mice (b), but not Slc6a8+/- mice [males: main effect of gene

ur na

(F(1,42)=40.44, p<0.0001); female: gene × treatment interaction (F(1,51)=4.78, p<0.05)]. Individual data points in (a) and (b) are the means per group for each trial. (c) Slc6a8-/y and (d) Slc6a8+/- mice in the SHOCK group had decreased latencies to escape relative to wildtype mice in SHOCK group [males: main effect of gene (F(1,22)=6.15,

Jo

p<0.05); trial × gene interaction (F(29,638)=2.55, p<0.0001); females: main effect of gene (F(1,29)=5.37, p<0.05)]. n=12-16/group, * gene effect, p<0.05; † treatment effect, p<0.05

Figure 4. PCA Challenge. Neither baseline locomotor activity nor activity in response to PCA were affected. PCA dosed at (a) 1 mg/kg, n=7-8/group, and (b) 2.5 mg/kg, n=8/group, failed to produce hyperactivity, but (c) 5 mg/kg produced robust hyperactivity

28

(F(1,48)=50.58, p<0.0001) in both groups, but no difference in distance travelled. n=13/group. Figure 5. Neurotransmitter and metabolite content. (a-c) HIPPOCAMPUS: Slc6a8-/y mice have increased 5-HIAA concentration (t22=2.384, p<0.05; b), but unchanged 5-HT and norepinephrine concentrations. (d-i) STRIATUM: Slc6a8-/y have increased 5-HIAA (t22=2.817, p<0.01, 1 knockout outlier removed; e), dopamine (t20=2.493, p<0.05, 1

of

outlier per group removed; g), and increased 3-HVA (t21=2.246, p<0.05, 1 knockout

ro

outlier removed; i) content relative to wild type controls. However, 5-HT concentration is unchanged (d). n=11-13/group, * p<0.05

-p

Figure 6. TPH-2 and MAO content in hippocampus. (a) Slc6a8-/y mice have a greater

re

concentration of TPH-2 in the hippocampus relative to Slc6a8+/y controls (t21=1.872, p<0.05, 1 wildtype outlier removed; (b) Representative composite of western blots

lP

showing TPH-2 (56 kDa, green band) and α-tubulin (~50 kDa, red band). (c) Slc6a8-/y mice also have a greater concentration of MAO in the hippocampus relative to WT

ur na

controls (t23=1.93, p<0.05). (d) Representative composite of western blots showing MAO

Jo

(60 kDa, green band) and β-actin (45 kDa, red band). n=10-13/group, * p<0.05

29

Figure 1

a

b 150

20

10

0

100

0

Slc6a8+/y Slc6a8-/y Slc6a8+/+ Slc6a8+/-

d 40

*

20

re

20 10

10

Slc6a8+/y Slc6a8-/y Slc6a8+/+ Slc6a8+/-

Jo

ur na

0

30 Head dips

30

-p

40

lP

Open quadrant entries

Slc6a8+/y Slc6a8-/y Slc6a8+/+ Slc6a8+/-

ro

c

50

of

Time in open (s)

Latency to first entry (s)

30

0

Slc6a8+/y Slc6a8-/y Slc6a8+/+ Slc6a8+/-

30

80 60 40

30

10

2

3

40

4

5

Slc6a8+/+

10

1

2

Jo

0

ur na

Total mobility (s)

Slc6a8+/-

3

Minute

4

5

e

Total immobility (s)

1

Minute

20

ro

50

Slc6a8+/y Slc6a8-/y

40 30

re

20

30

c Total immobility (s)

Slc6a8+/y Slc6a8-/y

lP

Total mobility (s)

d

Slc6a8+/y Slc6a8-/y Slc6a8+/+ Slc6a8+/-

40

0

of

20 0

b

*

-p

a

Latency to immobility (s)

Figure 2

20 10 0

1

2

3

4

5

Minute 40

Slc6a8+/+ Slc6a8+/-

30 20 10 0

1

2

3

Minute

4

5

31

Figure 3

b

No Shock Shock

No Shock Shock

18

12

*

6

0

24

18

12

6

0 2

4

6

8 10 12 14 16 18 20 22 24 26 28 30

2

4

6

of

24

Escape Latency (s)

Escape Latency (s)

a

8 10 12 14 16 18 20 22 24 26 28 30

Trials

12

re

12

18

No Shock Shock

-p

24

No Shock Shock

18

ro

d

24

Escape Latency (s)

Escape Latency (s)

c

Trials

6

6

4

6

8 10 12 14 16 18 20 22 24 26 28 30

Trials

ur na

No Shock Shock

*

20

10

Jo

Escape Latency (s)

e 30

0

Slc6a8+/y

*

*

Slc6a8-/y

f

Escape Latency (s)

2

lP

*

0

0

2

4

6

8 10 12 14 16 18 20 22 24 26 28 30

Trials No Shock Shock

30

*

*

20

10

0

Slc6a8+/+

Slc6a8+/-

*

32

Figure 4

a Slc6a8+/y Slc6a8-/y

Saline

2000 PCA

1500

of

1000 500 0

b

15 30 45 60 75 90 105 120 135 150 165 180

Minutes Saline

-p

Slc6a8+/y Slc6a8-/y

1500

re

PCA

lP

1000

500

0

c

15 30 45 60 75 90 105 120 135 150 165 180

ur na

Distance travelled (cm)

2000

Minutes

Saline

1500

1000

Slc6a8+/y Slc6a8-/y

PCA

Jo

Distance travelled (cm)

2000

500

0

ro

Distance travelled (cm)

2500

15 30 45 60 75 90 105 120 135 150 165 180

Minutes

33

Figure 5

a

b

Serotonin

2000

pg/mg

pg/mg

6000 4000

1000

1000

500

2000 0

Slc6a8+/y Slc6a8-/y Slc6a8+/+ Slc6a8+/-

d

e

Serotonin

Slc6a8+/y Slc6a8-/y Slc6a8+/+ Slc6a8+/-

f

5-HIAA

8000

3000

0

Slc6a8+/y Slc6a8-/y Slc6a8+/+ Slc6a8+/-

Norepinephrine

of

pg/mg

*

8000

3000

Norepinephrine

1500

10000

4000

0

c

5-HIAA

600

*

4000

1000

200

g

h

Dopamine

DOPAC

* 15000 10000

+/y

Slc6a8

2000

-/y

+/+

Slc6a8

+/+

Slc6a8

Slc6a8

+/-

3-HVA

*

1500 1000

500

5000

Slc6a8

+/y

Slc6a8

-/y

Slc6a8

+/+

Slc6a8

+/-

0

Slc6a8

Jo

ur na

0

1000

lP

pg/mg

pg/mg

20000

Slc6a8

i

re

1500

25000

0

Slc6a8+/y Slc6a8-/y Slc6a8+/+ Slc6a8+/-

pg/mg

0

Slc6a8+/y Slc6a8-/y Slc6a8+/+ Slc6a8+/-

-p

2000 0

400

ro

pg/mg

pg/mg

pg/mg

6000 2000

+/y

Slc6a8

-/y

Slc6a8

500

+/+

Slc6a8

+/-

0

Slc6a8

+/y

Slc6a8

-/y

Slc6a8

+/-

34

Figure 6

a

b 15

Slc6a8-/y

75 kDa

10

TPH-2 α-tubulin

50 kDa 37 kDa

5

0

Slc6a8+/y Slc6a8-/y Slc6a8+/+ Slc6a8+/-

d

ro

c

Slc6a8+/y

20

75 kDa 50 kDa

10

re

37 kDa

5

Slc6a8+/y Slc6a8-/y Slc6a8+/+ Slc6a8+/-

Jo

ur na

0

Slc6a8-/y

-p

* 15

lP

arbitrary units

Slc6a8+/+ Slc6a8+/-

of

arbitrary units

Slc6a8+/y

*

Slc6a8+/+

Slc6a8+/-

MAO ß-actin