Wild-type (WT), Slc6a8^Y389C knock-in [KI, (Duran-Trio et al., 2021)] heterozygous (Slc6a8^x/xY389C) and homozygous (Slc6a8^{xY389C/xY389C}) female rats (fWT, fHE, and fHO henceforce, respectively) were obtained from crossbreedings between fHE and WT or Slc6a8^Y389C KI males. Littermates of 3–4 months of age were used for all the experiments. Animals were genotyped and housed as previously described (Duran-Trio et al., 2021). All experiments were performed with approval of Canton de Vaud veterinary authorities (VD-3284) in accordance with the Swiss Academy of Medical Science and followed the ARRIVE Guidelines 2.0. Efforts were made to minimize stress and number of animals used.

In vivo proton magnetic resonance spectroscopy (¹H-MRS) at high magnetic field (9.4T) was performed in CNS as previously described (Duran-Trio et al., 2021). 4 fWT, 8 fHE and 4 fHO Slc6a8^Y389C female rats were used.

Tissue and liquids (plasma, urine) were collected as described (Duran-Trio et al., 2021). Measures of Cr and GAA were performed by liquid chromatography coupled to double mass spectrometry (LC/MS-MS) as described (Braissant et al., 2010; Duran-Trio et al., 2021). Measure of Crn was performed on a COBAS 8000 automate (Roche, Switzerland). Histology of muscle tissue was done as described (Duran-Trio et al., 2022).

Female rats were subjected to a battery of tests described in Duran-Trio et al. (2021; 2022): open field novel object (OFNO), elevated plus maze (EPM), light-dark box (LDB) and Y-maze spontaneous alternation. All tests were recorded with a videocamera and analyzed offline. A vaginal smear was done 2–3 h after each test (see below), and rats remained undisturbed 1–3 days between tests. To minimize circadian effects, animals were tested at the same time interval of the day.

For OFNO test, rats were introduced in a circular open arena (1 m diameter, black PVC) for 10 min (OF phase). Then, a novel object was located in the middle of the open arena and the rat was left undisturbed for other 5 min (NO phase). Spontaneous behavior was recorded with a videocamera. Average velocity, total distance moved, and time spent moving or not moving (considering thresholds of 2 and 1.75 cm/s, respectively) were calculated using EthoVision tracking software per each phase. Additionally, arena was virtually segmented in three concentric areas (“center” is the 12.5 cm radius region in the center of the arena, “wall” is the external region of 12.5 cm radius from the wall, and “intermediate” is the resting region between them) and cumulative duration spent in each one was analyzed with EthoVision. Index were calculated as the relative difference between NO and OF phases [Index = (NO – OF)/OF]. “Index Dur” is from cumulative duration (in %) spent in each region and phase, and “Index Mov” is from cumulative duration moving (in %) in each phase.

OF phase (10 min) recordings were used for the analysis of grooming behavior, rearing supported (standing on hind limbs and one or two superior extremities) and unsupported (no superior limb used). Such behaviors were hand scored with The Observer XT software.

The elevated plus maze (EPM) test consists of two opposing open arms (45 × 10 cm, 16 lux) and two opposing enclosed arms (45 × 10 × 50 cm, 4 lux) that extend from a central platform (10 × 10 cm, 11 lux), elevated 65 cm above the floor. Rats were placed individually on the central platform facing an enclosed arm and were allowed to freely explore the maze for 5 min. The behavior of each rat was monitored using a videocamera, and the movements of the rats were automatically registered and analyzed with EthoVision^® tracking software. Entry into an arm was defined as entry of all four paws into one arm. Cumulative duration and number of times the animal spent in each type of arm were analyzed, as well as head dipping behavior (looking over the edge of an open arm or the central platform) and the latency to the first open arm.

The light-dark box apparatus consists of two compartments (40 × 40 × 50 cm each), one with black walls and floor, illuminated at 20 lux, and the other with white walls and floor at 160 lux. The two compartments are connected by a small opening (8 × 10 cm high) located in the center of the partition at floor level. During the test the animals were individually placed in the middle of the white compartment facing away from the opening of the dark box and were recorded with a videocamera for 5 min. The time spent in the black compartment, the number of entries into the white compartment and the latency to the black compartment were scored with EthoVision^® tracking software. Some videos were hand-scored blindly with The Observer™ software to check the automatic analysis.

Same test and procedure as previously described (Duran-Trio et al., 2021).

A tip of 1 ml with 200 μl of saline (0.9 % NaCl) was introduced in the vagina (no more than 0.5 cm deepness) of an immobilized rat (tail up to expose the genitals). The liquid was pipetted in and out 2–3 times, then extended on a slide and let dried in a heat pad. Hematoxylin-eosin staining was performed as described (Duran-Trio et al., 2022). Using an OLYMPUS microscope, estrous phases were identified according to Cora et al. (2015). For the analysis, 3 factors were considered (“diestrous”, “estrous”, and “rest” where proestrous and metestrous were grouped together).

Images of stained muscles were taken on an OLYMPUS microscope; cross-sectional areas and minimum Feret diameter were obtained with ImageJ (Duran-Trio et al., 2022).

All statistical analysis and graphs were conducted with R-3.5.1 (R Core Team, 2018). A Shapiro test was used to assess the normality of each sample. Mann-Whitney test with Bonferroni correction was used for pair comparisons among genotypes when normality was rejected. For normal distributions, we used one-way analysis of variance (ANOVA) followed by Tukey post hoc test for pair comparisons among genotypes, or t-test to address the significance between a genotype with respect to the value expected by chance.

For behavioral tests, generalized linear mixed models blocking estrous cycle [diestrus, estrous, rest=(proestrous+metestrus)] as random factor were used to analyze the differences among genotypes [package nlme (Pinheiro et al., 2015)], and obtained coefficients (mean and standard error of the mean) were used for graphs. Graphs were done using ggplot2 package (Wickham, 2016). P-values are reported in figure legends, and statistical significance was considered at P < 0.05.

Both Slc6a8^Y389C fHE and fHO female rats are viable and fertile, although the later ones showed reduced fertility (giving only half of the offspring) when crossed with Slc6a8^Y389C KI males.

Reduced body weight gain is characteristic of CTD male patients and rodent models of CTD, including the KI rat males (Baroncelli et al., 2014; Duran-Trio et al., 2021; van de Kamp et al., 2013; Miller et al., 2019; Skelton et al., 2011; Stockebrand et al., 2018). In comparison with fWT, Slc6a8^Y389C fHE did not show significant differences in body weight gain along time, while Slc6a8^Y389CfHO presented significant less body weight gain along time from 6 weeks-old on (peri-adolescence) (Figures 1A, B).

Due to CrT deficiency in kidney, Slc6a8^Y389C KI male rats present elevated Cr/Crn (as CTD male patients) and GAA/Crn ratios in urine in comparison with WT male rats (Duran-Trio et al., 2021). Given that fHO rats also lack functional CrT, we expected that Cr will be lost in the urine too, thus showing elevated urinary Cr/Crn ratio. On the other side, cells from fHE rats will or will not express a functional CrT, depending on which X-chromosome is silenced. Therefore, taking into account the mosaics' variety, a broad spectrum from normal to elevated Cr levels in urine would be theoretically expected.

Indeed, fHO rats presented a significant increase in both urinary Cr and GAA concentrations and a significant decrease of Crn levels (+600%, +300% and −67% vs. fWT, respectively; Supplementary Figure 1A), resulting in significant elevated urinary Cr/Crn and GAA/Crn ratios in comparison with fWT (× 148 and × 16, respectively; Figure 1C and Supplementary Figure 1A).

On the other hand, compared to fWT, fHE rats presented significant elevated urinary Cr levels (x7) and slight but not significant increase in GAA and Crn concentration (Supplementary Figure 1A), resulting in a significant increase in urinary Cr/Crn (x11) ratio but no significant changes in GAA/Crn ratio (Figure 1C and Supplementary Figure 1A).

In comparison to fWT rats, both fHE and fHO rats exhibited significant changes in plasmatic Cr and GAA levels, decreasing the first one and increasing the later one (Supplementary Figure 1A). However, plasmatic Crn levels only decreased in fHO (Supplementary Figure 1A).

CTD male patients are characterized by brain Cr deficiency and are diagnosed using ¹H-MRS. This is also the case in CTD rodent models, including the Slc6a8^Y389C rat (Duran-Trio et al., 2021). High magnetic field 9.4 T ¹H-MRS spectra from cerebellum (Figure 1D) and hippocampus (data not shown) systematically showed much lower peaks corresponding to Cr and PCr in Slc6a8^Y389C fHO compared to those of fWT ones, while a broad variety of peaks heights was observed in fHE spectra. Levels of Cr and PCr together (Cr+PCr) were significantly reduced in fHE rats compared to those of fWT; and those of fHO (−73% and −84% vs. fWT in cerebellum and hippocampus, respectively) to those of fWT as well as to those of fHE (Figure 1E).

Taking advantage of the high resolution 9.4T ¹H-MRS measures to quantify potential metabolic changes in the brain of Slc6a8^Y389C rats, we observed that fHO, but not fHE, presented significant differences in some metabolites' concentrations in comparison with those of fWT ones (Supplementary Figure 1B). Gln, Glu, and Gln+Glu were increased in cerebellum, while NAA levels were increased in both cerebellum and hippocampus, and NAA+NAAG only augmented in hippocampus.

Muscle Cr levels significantly decreased while GAA levels significantly increased in fHO rats compared to those of fWT or fHE ones (Figure 2A). However, only muscle GAA levels from fHE rats significantly differed from those of fWT ones.

Furthermore, myocytes from fHO rats exhibited significantly smaller minimum Feret diameter and cross-sectional area compared to those of fWT or fHE ones (Figures 2B, C). Myocytes from fHE rats presented similar size than those of fWT rats.

Locomotor activity was evaluated during free exploration in an open field. When compared to fWT, fHO rats showed significant less distance moved, less cumulative duration moving, less cumulative duration doing supported rearing and less frequency of supported and unsupported rearing in comparison with fWT rats (Figures 2D, E). In contrast, fHE rats showed similar performance as fWT rats.

Our previous findings showed that Slc6a8^Y389C KI rat males exhibited bad performance in a working memory and attention task (the Y-maze alternation test) and increased grooming behavior (Duran-Trio et al., 2021).

However, neither fHE nor fHO rats showed significant differences with respect to fWT rats in the total duration, frequency or latency of grooming behavior evaluated in similar conditions (Figure 3A).

Additionally, in the Y-maze task, there were no significant differences among genotypes in the number of entries, although fHO rats showed a tendency to reduce it and significant less distance moved compared to both fWT and fHE rats. Noteworthily, fHE and fHO rats showed no significant differences in percentage of alternation in comparison with fWT rats (Figure 3B), suggesting for all genotypes a good performance in the task (since their values significantly differed from the value expected by chance: 100/3).

We observed that, at the onset of adolescence (7–8 weeks-old), both fHE and fHO rats significantly started displaying abnormal behavior (Figure 3C) when handling them in comparison with fWT rats (around 6% for fWT, while 19% and 38% for fHE and fHO, respectively). They screamed, trying to flee, increasing locomotion and jumping out. At early adulthood (11 weeks-old), both fHE and fHO needed longer habituation time (doubling the number of days in some cases) to handle them without signs of stress before performing the behavioral tests. Additionally, we also observed that both mutant female rats, specially fHO rats, were more prone to bite both other rats in the same cage and human researchers.

In order to evaluate rats' behavioral response to stressors, anxiety-like behavior and locomotion were assessed in different tests that involve 3 different types of stressors, including novelty (OFNO), openness (EPM) and light (LDB).

In the OFNO test, there were no significant differences among genotypes in the OF phase except for fHO rats who moved less distance in comparison with fWT and fHE rats (Figure 2D and Supplementary Table 1). Due to their natural aversion to open spaces, rats explored freely the open field spending longer time close to the wall (nearly 70% of the time, 3% on the arena center). However, after locating the novel object in the center of the arena (NO phase), fWT rats decreased their time spent in the wall zone (to 65%) by increasing their time exploring the novel object in the center (to 5%) and surroundings (Figure 3D and Supplementary Table 1). In contrast, fHE rats significantly raised their time spent in the wall zone (to 85%) and reduced it nearby the novel object. Additionally, fHE rats significantly decreased their time moving and the total distance moved in comparison with fWT females. In contrast, fHO females did not show significant differences with respect to fWT rats. These results indicate that fHE females present increased anxiety-like behavior to novelty.

In the EPM test, due to rodents' natural aversion to open spaces, fWT rats stayed longer time in the closed arms than in the opened arms (60 vs. 19% of the time). Interestingly, although both fHE and fHO rats did not show significant differences in the duration spent in each compartment in comparison with fWT ones (Figure 3E), they significantly exhibited an opposite pattern in the exploration of closed and open spaces: in comparison with fHE rats, fHO rats spent longer time in the closed arm and less (around the half of the time) in the open arm and doing head dipping. These results indicate that fHE and fHO rats display normal anxiety-like behavior in this test, being, respectively, at the anxiolytic and anxious extremes of a wild-type phenotype.

In LDB test, due to rodents' natural aversion to illuminated places, fWT rats spent only around 25% of the time in the light compartment and 75% in the dark one. A similar pattern was presented by fHO rats, but fHE rats stayed significantly longer time in the light compartment and moved with higher velocity than fWT and fHO rats (Figure 3F). These results indicate that fHE rats present altered behavioral response and hyperactivity to light as stressor.

In summary, both mutant Slc6a8^Y389C female rats present behavioral abnormalities in response to stressors, namely increased or decreased anxiety-like behavior with or without hyperactivity, depending on the trigger and the zygosity.