A new rat model of creatine transporter deficiency reveals behavioral disorder and altered brain metabolism

Creatine is an organic compound used as fast phosphate energy buffer to recycle ATP, important in tissues with high energy demand such as muscle or brain. Creatine is taken from the diet or endogenously synthetized by the enzymes AGAT and GAMT, and specifically taken up by the transporter SLC6A8. Deficit in the endogenous synthesis or in the transport leads to Cerebral Creatine Deficiency Syndromes (CCDS). CCDS are characterized by brain creatine deficiency, intellectual disability with severe speech delay, behavioral troubles such as attention deficits and/or autistic features, and epilepsy. Among CCDS, the X-linked creatine transporter deficiency (CTD) is the most prevalent with no efficient treatment so far. Different mouse models of CTD were generated by doing long deletions in the Slc6a8 gene showing reduced brain creatine and cognitive deficiencies or impaired motor function. We present a new knock-in (KI) rat model of CTD holding an identical point mutation found in patients with reported lack of transporter activity. KI males showed brain creatine deficiency, increased urinary creatine/creatinine ratio, cognitive deficiency and autistic features. Slc6a8xY389C KI rat fairly enriches the spectrum of CTD models and provides new data about the pathology, being the first animal model of CTD carrying a point mutation.


Contribution of authors
LDT and OB designed the study and wrote the manuscript; LDT and GFP handled the animals and collected liquids; CR and PAB performed Cr, GAA and Crn measures; LDT, JG and CS designed and performed the behavioral tests; CC and DS performed the 1 H-MRS measures; LDT analyzed the data; all authors reviewed and corrected the manuscript.

INTRODUCTION
Creatine (Cr) is a nitrogenous organic acid used as phosphate energy buffer to recycle ATP by the Cr-PCr system; Cr is also one of the main osmolytes of the brain, and suggested to play a role as neuromodulator or neurotransmitter 1,2 . Cr is taken from the diet or endogenously synthetized in a two-step pathway by the enzymes AGAT (arginine-glycine amidinotransferase) and GAMT (guanidinoacetate methyltransferase), and specifically transported into the cells by SLC6A8 (also called CRT, CreaT or CT1, henceforth CrT) 2 .
Guanidinoacetate (GAA) is synthetized from arginine (Arg) and glycine (Gly) in the first step, and it is methylated generating Cr in the second step. Cr breaks down spontaneously in creatinine (Crn).

Deficits in endogenous synthesis or transport of Cr cause Cerebral Creatine Deficiency
Syndromes (CCDS) [3][4][5][6] . CCDS are characterized by brain Cr deficiency, intellectual disability with severe speech delay, behavioral abnormalities such as attention deficits and/or autistic features, and seizures for GAMT and SLC6A8 deficiencies 7 . Among CCDS, Cr transporter deficiency (CTD) is caused by loss-of-function mutations in the SLC6A8 gene (for a scheme of the human locus with the position of the mutations, see Figure 1A). CTD is a X-linked gene disorder with an estimated prevalence of about 2% of males with intellectual disability with, unlike the other two CCDS, no efficient treatment so far 1,7,8 .
Brain-specific KO mice of CrT have shown additional and somehow different phenotypes.
Deletion of exons 2-4 11 or 5-7 13,18 in CNS nestin-positive cells exhibited brain Cr deficiency but normal Cr levels in periphery. Brain CrT KO of exons 2-4 mice maintained the same learning and memory deficits as the corresponding ubiquitous KO, plus hyperactivity and normal body weight 11 . In contrast, brain CrT KO of exons 5-7 mice showed a rescue in the general phenotype (no stereotypies, normal memory performance and body weight) but declined at early age, correlating with some brain alterations related with aging 13,18 . Deletion of CrT exons 2-4 in CaMKII-positive cells (excitatory neurons) 15 revealed similar phenotype as the nestin-targeted KO mouse, while targeting only DAT-positive cells (dopaminergic neurons) exhibited hyperactivity 19 .
All the accumulating knowledge using these various mouse models highlights the importance of the Cr transporter in brain function and physiology, and allows a better understanding of SLC6A8 deficiency. However, CTD patients do not always present the same phenotype 17 .
Further studies are thus needed to fully understand CTD pathology, and to allow, through the use of various complementary animal models, the best translation to human pathology 20 .
As an alternative to mouse models, rats have a number of potential advantages over mice. In particular, they are closer to human physiology, biochemistry and behavior in comparison with mice 21,22 . In order to increase the fundamental knowledge and to increase the relevance of preclinical studies with complementary preclinical models, we generated a novel rat model of CTD: the Slc6a8 Y389C knock-in (KI) rat line. Slc6a8 Y389C KI rat holds a point mutation found in CTD patients which was reported to completely abolish Cr transporter activity 7 .
Here we show the first analysis of Slc6a8 xY389C/y rat KI males as model of CTD, being the first KI animal model of CTD.

Generation, genotyping and housing of the Slc6a8 Y389C KI rat strain
Based on the patient SLC6A8 c.1166A>G; p.(Tyr389Cys) point missense mutation reported as completely abolishing Cr transporter activity 17 (Figure 1 A) mm, 2 averages, 256x256 image matrix). The static magnetic field homogeneity was adjusted using first and second order shims by FAST(EST)MAP leading to water linewidths of 10-12Hz for hippocampus and striatum and 14-16Hz for cerebellum 23 . Localized spectra were acquired with SPECIAL spectroscopy sequence (TE=2.8 ms, TR=4 s, 160 averages) to obtain the specific neurochemical profile in the 3 mentioned VOIs 24,25 . Metabolite concentrations were estimated using LCModel software 26

CSF, urine and plasma collection
The same day of the sacrifices, 6 WT and 7 KI rats aged of 3 months were anesthetized with 4% isoflurane under 70% compressed air and 30% O2. Fur around the base of the head and the dorsal part of the neck was shaved with an electric shaver. A little incision was made in medial position and longitudinally through the skin and muscles at the base of the skull. The cavity was maintained opened using little separators and cleaned of blood using cotton swabs.
Carefully, puncture in cisterna magna was made to collect CSF. After CSF collection, the rat was positioned in supine position and the abdominal cavity was opened to collect urine from the bladder. Finally, the thoracic cavity was opened and blood was collected transcardially with heparin (5000 u/ml) and centrifuged at 2710 x g for 7 min. Supernatant (plasma) was transferred in a new tube. Liquids (CSF, urine and plasma) were stored at -80 ºC for further analysis.

Measure of creatine, guanidinoacetate and creatinine in plasma, urine and CSF
Cr and GAA were quantified by liquid chromatography coupled to mass spectrometry (LC-

Behavioral tests
Two cohorts of males were used to do the behavioral tests: grooming and Y-maze for spatial memory test in one; novel object recognition (NOR), Y-maze for spontaneous alternation and Social preference/Social memory (SP-SM) tests for the other one. WT and KI were littermates in each cohort and the set of tests began at 12 weeks of age.

Grooming
Rats were introduced in a round (1m diameter) open arena with black walls and floor made of black PVC for 10 min while spontaneous behavior was recorded with a videocamera.
Grooming behavior was scored blindly with the Observer XT® software and cumulative duration and frequency of such behavior were analyzed. 10 WT and 10 KI males were used.

Y-maze for spatial memory test
A Y-shape maze of 50 cm arm length, grey walls and floor made of grey PVC in conditions of 4 lux in each arm was used. For the learning phase, the rat was introduced in the Y-maze for 10 min with one arm closed. After 4 hours, the rat was located in the same starting arm and was allowed to explore all the arms for 5 min. EthoVision® software was used to track the time spent in each arm. Exploration time (tN + tF, in seconds) and Recognition Index ((tN -tF)/(tN + tF)) were calculated from the cumulative time spent in the familiar (tF) and the new (tN) arms. 9 animals per group were used.

Novel Object Recognition (NOR) test
An arena of 57x36cm with grey walls and floor made of grey PVC in 4 lux conditions was used. For habituation, the rat explored the empty arena for 5 min. On next day, the rat was introduced in the arena with two identical objects for 5 min in the learning phase. After 4 hours, the rat was allowed to explore again the arena for 5 min but one of the objects was changed for a different one. Both, new and familiar objects and the right and the left sides were randomized. All phases were recorded with a videocamera and exploratory behavior was scored blindly with the Observer TX® software. The cumulative duration times for exploring familiar (tF) and new object (tN) in testing phase were used to calculate Exploration Time (in seconds, tN + tF) and Recognition Index ((tN -tF)/(tN + tF)). 8 WT and 9 KI males were used.

Y-maze spontaneous alternation
The rat was introduced in the abovementioned Y-maze and allowed to explore the three arms opened for 8 min. EthoVision® software was used to track rat position and calculate the number of total entries in all the arms (nº entries in arm A + arm B + arm C) as well as the percentage of alternation (100*frequency of alternations/(nº total entries -2)). 8 WT and 10 KI males were used.

Social Preference and Social Memory (SP-SM) tests
A three-chamber arena 80x34cm with grey walls and floor made of grey PVC with 10 lux in the center was used for SP-SM test. A transparent cylinder made of PVC, 13 cm in diameter and with openings, was positioned in each lateral chamber. During habituation, a juvenile rat was located in one cylinder while an object was positioned in the other one, and the experimental male rat was allowed to explore only the center chamber for 5 min. During the social preference phase, doors were opened and the rat could explore all the chambers (with the juvenile rat or the object) for 10 min. After 2 hours, the object was changed for another juvenile, and the experimental rat was allowed to explore both new and familiar juveniles for 10 min. Relative positions of the items (object and juveniles) were randomized. All phases were recorded with a videocamera and exploration behavior towards the object or the juveniles was scored blindly with the Observer®. Cumulative duration time exploring each item, the number of visits (frequency) per each item, and the first item to be explored in each phase were taken into account for the analysis. 9 WT and 10 KI were used.

Statistical analysis and graphs
All statistical analysis and graphs were conducted with R-3.5.1 28 . A Shapiro test was used to assess the normality of each sample. To address the significance between WT and KI groups or between each group with respect to the value expected by chance, Mann-Whitney tests were performed when normality was rejected while, for normal distributions, we used Bartlett tests to evaluate equality of variances and t-tests for equality of the means.
To calculate the log10 of the ratio KI/WT concentrations of different metabolites in brain, a pseudocount (+0.001) was added to all concentrations before doing the ratio. Graphs were done using ggplot2 package 29 , while heatmaps were generated with gplots 30 .
In the analysis of first decision for the SP-SM experiment, we evaluated different questions.
First, we asked if there was a significant bias in the decision of exploring first one of the two possibilities. We modelled this question considering a Bernouilli variable with probability q=0.5 (no preference), and estimating the probability of selecting one of the two choices with a binomial distribution of parameters q, N, k, where N is the number of animals tested and k the number of visits of one of the items. We tested if the preferred choice was significantly different than random performing a two-tail exact binomial test, or a one-tail exact binomial test when prior information existed (e.g. when a given behavior was observed for the WT and the preference of the KI was tested, or when the preference was known a priori from literature). Finally, we tested if the preferences of KI and WT were significantly different between them using a Barnard's CSM Test (exact.test, method=CSM, package Exact 31 ).

Generation of the Slc6a8 Y389C rat
To obtain a model as near as possible of the human CTD, a knock-in (KI) rat was generated with the SLC6A8 c.1166A>G; p.(Tyr389Cys) missense mutation found in human patients and reported with no functional activity of the transporter 17 (Figure 1). The obtained Slc6a8 Y389C rats were viable and bred normally, generating litters with the expected Mendelian ratios. systematically showed much lower peaks corresponding to Cr and PCr in Slc6a8 xY389C/y KI males compared to those of WT ones. Concentration of Cr and PCr in Slc6a8 xY389C/y KI males was 70-75% less than that of WT in every brain region that was measured (Figure 2 B).

Slc6a8 Y389C rat males present a decrease of Cr and an increase of GAA levels in plasma
We collected plasma and CSF from WT and Slc6a8 xY389C/y KI males at around 3 months of age to see whether Cr and GAA concentrations were altered. We show a significant decrease of Cr and a significant increase of GAA concentrations in plasma of KI males in comparison to those of WT males ( Figure 3A left and right panels, respectively). Cr and GAA levels in CSF from Slc6a8 xY389C/y KI males at around 3 months of age showed no differences in comparison to those of WT males ( Figure 3B left and right panels, respectively).

Slc6a8 Y389C rat weight gain is slowed
KI males (Slc6a8 xY389C/y ) presented a significant reduced body weight gain just after weaning, as compared to WT males, showing for instance 56% of the body weight of their WT littermates at 14 weeks of age (312.7 ± 8.6 g versus 524.6 ± 12.1 g, p<0.001, Mann-Whitney test) (Figure 4).

Other brain metabolic changes in Slc6a8 Y389C rats
As described above, Slc6a8 Y389C rats showed brain Cr deficiency demonstrated by 1 H-MRS in every CNS region measured (Figure 2A-B). We further took advantage of our high resolution 9.4T 1 H-MRS measures to quantify potential metabolic changes in the brain of Slc6a8 xY389C/y KI males. Interestingly, Gln, NAA, and the sum of NAA and NAAG (NAA+NAAG) concentrations were significantly increased in all the brain regions analyzed while other metabolites changed their levels only in specific brain regions (Figure 5 A and   B). In hippocampus, GABA, Glu, Tau and total choline (GPC+PCho) were significantly increased. In striatum, Ins increased while Lac decreased significantly. Interestingly, Ins tended to change in all brain regions although in different directions (Figure 5 B).
CTD patients present cognitive deficiency and behavioral troubles such as autistic-like behaviors or attention deficits. In order to analyze if our Slc6a8 Y389C rat line is a good model for neurological symptoms of CTD, we performed several behavioral tests in relation with those symptoms.

Slc6a8 xY389C/y KI males show similar performance than WT ones on NOR and Y-maze for spatial memory tasks
To assess cognitive deficiency-like symptoms, we evaluated learning and memory performance in two different tasks. Y-maze test evaluates short-term spatial memory while Novel Object Recognition (NOR) test measures declarative memory. In Y-maze, both recognition index and total exploration time of the arms were similar between genotypes ( Figure 6 A and B, respectively). In NOR, Slc6a8 xY389C/y KI males showed similar recognition index than WT littermates (Figure 6 C), although total exploration time of the objects was slightly higher in the testing phase (Figure 6 D) and significantly higher in the learning phase (+16% from WT, p=0.043). This increase in exploration time without any improvement in learning and memory performance may indicate attention deficits and/or impairment in working memory capacity.

Slc6a8 xY389C/y KI males present deficits in working memory
Working memory affects learning and memory 32 . Additionally, to assess attention deficit in rodents, a task evaluating working memory is typically used. Thus, we carried out Y-maze for spontaneous alternation test in a different cohort of animals. Percentage of alternation index was used to compare working memory performance between genotypes and each genotype with the value of a random exploration (33.3% of alternation). While there were no differences in the number of entries between genotypes (Figure 7 A), the percentage of alternation for Slc6a8 xY389C/y KI males was significantly reduced in comparison with that of WT males, having similar value than the expected by chance (Figure 7 B). This result shows impaired working memory in KI males.

Slc6a8 xY389C/y KI males present more repetitive self-stimulating behavior
To analyze whether Slc6a8 xY389C/y KI males present autistic-like behavior, we decided to evaluate the presence of repetitive self-stimulating behaviors like grooming. In a period of 10 min in an open field, KI males spent significantly more time doing grooming than WT males, and with higher frequency (Figure 7 C and D, respectively).

Slc6a8 xY389C/y KI males show a tendency in decreased sociability
Decreased sociability is one core feature of autism spectrum disorders. We used Social Preference test (SP) to evaluate the general sociability and Social Memory test (SM) to evaluate the interest in social novelty. For this analysis, we used the cumulative duration of the rat exploring an item (juvenile, object, new or familiar juvenile, Figure 8 A, B), the number of the visits for each item (frequency, Looking at the two phases together, Slc6a8 xY389C/y KI males systematically presented a tendency in reducing the cumulative duration exploring the juveniles (around -17.5% less than WT males) but not the object (+7.6%). These little changes of KI males resulted in a nearly significant reduction of the social exploration time along the two phases (Figure 8 B, p=0.051 two-tail t-test). Furthermore, the number of visits of KI rats towards the juveniles across the two phases was in general significantly lower than those of WT littermates but similar towards the object (Figure 8 C). As expected, we also observed that there were more WT animals going to explore firstly the juvenile (7 out of 9) instead of the object, being marginally significant with respect to the random expectation (p=0.090 one-sided exact binomial test, Figure 8 D green panel). In contrast, a lower percentage of KI animals decided to visit the juvenile in the first place (7 out of 11), being not significantly higher than the number expected by chance (p=0.27 one-sided exact binomial test, Figure 8 D green panel).
In summary, Slc6a8 xY389C/y KI males presented social preference but with a systematic pattern of reduced sociability when compared to their WT littermates.
Regarding the social novelty paradigm, WT and Slc6a8 xY389C/y KI males spent similar time with the new and the familiar juveniles, showing no differences between the two genotypes (Figure 8 A). Interestingly, while the number of WT males exploring firstly the new juvenile was similar than the number expected by chance (5 out of 9 rats, p=1.00 two-sided exact binomial test, Figure 8 C yellow panel), the number of KI males tended to be lower (3 out of 11 rats, p=0.23 two-sided exact binomial test). These data may point towards an aversion of KI males in establishing the first contact with a social novelty.

DISCUSION
We present here the first analysis of a new in vivo model of CTD, the Slc6a8 Y389C knock-in rat holding an identical patient's missense mutation abolishing CrT activity 17 . Slc6a8 xY389C/y KI rat males are brain creatine deficient, and exhibit increased Cr/Crn urine ratio as well as neurological symptoms similar to those of CTD patients, such as cognitive deficiency and some autistic features.
Using similar techniques as in the clinic, we first analyzed the brain metabolic profile of The increase in urine Cr levels reveals a functional deficiency in the reuptake of Cr from the primary urine, which is normally done at the renal proximal tubular cells by CrT 33 . Given that plasma Cr is being filtered and lost in the urine due to CrT deficiency, we could expect that Cr levels in plasma were diminished and those of GAA, the precursor of Cr, were increased in an attempt to compensate Cr levels by AGAT feedback loop of up-regulation due to Cr deficiency (for this mechanism in periphery 2,34,35 or for CNS 36 ). Indeed, a tendency to decreased Cr levels and significant increased GAA were observed in the plasma of Slc6a8 xY389C/y KI males. This finding is in agreement with the decreased levels of Cr also observed in the ubiquitous KO mouse males with exons 2-4 deleted 16 . These data contrast however with the low number of patients with available measurements of Cr and GAA in plasma showing normal or mildly elevated Cr levels as well as normal GAA 7,17 . Further studies are needed to unravel the possible causes of these differences between human and rodents.
Furthermore, we present the first measurement of the GAA/Crn ratio in urine of an in vivo CTD model. This ratio was strongly and significantly elevated in Slc6a8 xY389C/y KI males, suggesting that GAA/Crn ratio in urine may also be a valid biochemical marker for CTD when is found together with increased Cr/Crn ratio. This result might help to bring attention towards a CTD diagnosis in patients with suspected GAMT-D (where increased GAA levels in urine are characteristic, although with a low or slightly low Cr/Crn ratio 7 ). New data from patients will help to confirm whether this double marker would be useful for CTD diagnosis.
Whether GAA is also re-absorbed from primary urine in proximal tubular cells of kidney through transport by CrT (as we have shown in brain cells 37 ) remains to be determined.
Indeed, while Cr was decreased in plasma and increased in urine of Slc6a8 xY389C/y KI males demonstrating CrT deficiency, GAA was increased in plasma but normal with a tendency to increase in urine.
Consistent with CTD patients 17 and all the ubiquitous CrT KO mouse models 12,14,16 , Slc6a8 xY389C/y KI males presented lower body weight. We observed this difference already after weaning, but taking into account the degree of significance, it is likely that it extends from previous time points. This body weight gain issue is in agreement with reported early signs of CTD 38 .
Although Cr levels in plasma and in brain were lower in Slc6a8 xY389C/y KI males than in their WT littermates, Cr levels in the CSF were mildly higher or similar to those of WT ones. This is consistent with what was observed in CTD patients 7 , confirming that CSF does not always reflect what is happening in brain parenchyma, and that CSF can also be a useful diagnostic sample for the differential diagnosis between CTD and GAMT deficiency, in which Cr is decreased 7 . Interestingly, while WT males showed higher levels of Cr in plasma than in CSF, Slc6a8 xY389C/y KI males exhibited similar levels in both liquids (60-70 µmol/l). Altogether, these data support the fact that the Cr pool in CNS does not fully depend on the peripheral one, owing to the low permeability of blood-brain barrier for Cr and to brain endogenous synthesis of Cr 9,37,39,40 .
In every brain region analyzed, the strongly decreased Cr levels in brain parenchyma from Slc6a8 xY389C/y KI males were accompanied by significant increases in Gln, NAA and NAA+NAAG. Given that these molecules are important osmolytes in CNS [41][42][43] , and that the magnitudes of both changes tend to be similar, it might hint at an osmolyte compensation mechanism by Gln, NAA and/or NAAG for the brain Cr deficiency. Other metabolites with a role in osmoregulation are taurine, myo-inositol (Ins) or choline [44][45][46] . Interestingly, the concentration of these metabolites change in some of the regions analyzed as well (see below). We have recently shown a similar potential mechanism of osmolytes compensation involving the same molecules in another pathological condition of CNS (hepatic encephalopathy 24 ). As Gln, NAA and NAAG play other important roles in CNS, a line for future research is to establish whether their changes only compensate Cr deficiency osmotically, or are truly pathological.
Other metabolites changed differentially depending on the brain region analyzed. In hippocampus, neurotransmitters such as GABA, Glu and Tau increased together with total choline, a precursor for another neurotransmitter (acetylcholine), membrane phospholipids and, (via betaine) the methyl group donor S-adenosylmethionine which is involved in the synthesis of Cr from GAA 47 . Gln is a precursor for Glu, which is a precursor for GABA.
Transporters for GABA and Tau can transport Cr or GAA 39 . It would thus be possible that alterations in Cr and/or GAA levels might affect GABA and Tau homeostasis and their concentrations. As described above Tau and total choline in hippocampus might also participate to an osmolyte compensation mechanism under Cr deficiency. Overall, these metabolic changes may suggest that endogenous Cr synthesis may also be affected, together with disturbances in neurotransmission in the hippocampus of Slc6a8 xY389C/y KI males.
Hippocampus is involved in motivation, emotion, learning and memory, and these changes could affect its functions and also promote epilepsy in CTD patients. In striatum, Slc6a8 xY389C/y KI males showed an increase in Ins, osmolyte also involved in many other cellular processes. The decrease in lactate might reflect a change in striatal energy metabolism.
These alterations in brain metabolites correlate with changes in behavior and cognition. In comparison with their WT littermates, Slc6a8 xY389C/y KI males showed similar learning and memory performance in simple tasks about recognition and spatial memory (NOR and Ymaze for spatial memory tests) but failed in a working memory task (Y-maze for spontaneous alternation). Higher exploration times of objects in both phases of NOR test without any advantage in declarative memory might reflect a deficit in cognition related with attention and working memory as well. These negative results in learning and memory performance contrast with the data from the ubiquitous CrT KO mice with exons 2-4 and 5-7 deleted, where deficits were present although with milder phenotype in NOR test around the same age in the last one 13,16 . Interestingly, these negative results are consistent with data from CTD patients, as they are able to remember people, objects or locations but fail in the performance of more complex daily tasks and learning new abilities 17 A lot of evidence about working memory supports the relationship with learning and memory processes 32 as well as the significant impact on the performance of daily tasks 48 . Working memory capacity is a good predictor of learning outcomes 49 and general intelligence 50 .
Overall, our data may suggest that impairment in working memory, rather than deficits in declarative memory per se, is one of the primary components in the intellectual disability from CTD patients. Such hypothesis could open new avenues for alternative therapies to support working memory in order to improve learning in patients. Further studies are needed to address these questions.
Besides intellectual disability and attention deficits, autistic-like behavior is another common symptom in CTD patients. We evaluated the presence of two core features of autism spectrum disorder: repetitive self-stimulating behaviors (like grooming for rats) and sociability.
Slc6a8 xY389C/y KI males spent significantly more time in grooming and with more frequency, consistent with another CTD model 13 . Additionally, KI males appeared to be social but less than their WT littermates. This less social tendency is seen in the systematic pattern of spending less time with less frequency exploring the juveniles in all phases in contrast to exploring the object, and by having less tendency to first visit the juvenile instead of the object. Additionally, while WT males showed no preference to first visit any of the juveniles, KI males showed slightly more preference towards the familiar juvenile in the first place.
This result might reflect an aversion to establish the first contact with a social novelty, which is another feature of sociability present in autism. Interestingly, these results are in line with the autistic phenotype of the CTD patient sharing the same missense mutation as Further studies will be needed to unravel the nature of this difference.
In summary, this new Slc6a8 Y389C rat line is the first CTD model generated so far with one of the single nucleotide mutations described in CTD patients.      test recognition index (C) was similar between genotypes (p=0.97), although total exploration time (in seconds) (D) was slightly higher in KI males (p=0.064). 9 WT + 9 KI males were used for Y-maze test, 9 WT + 9 KI for NOR test; p-values were estimated with two-tail t-tests. n.s.: not significant. higher than the value expected by chance (p=0.39 one-tail t-test) while % of alternations in WT males were significantly higher (p=0.010 one-tail t-test). (C) KI males showed more time doing grooming (p=0.044) and (D) with more frequency than WT littermates (p=0.028, Mann-Whitney test). 8 WT and 10 KI males were used for Y-maze for spontaneous alternations, 10 WT and 10 KI for grooming behavior. By default, p-values were estimated with two-tail t-test. n.s.: not significant; *: p<0.05.