Deletion of the creatine transporter gene in neonatal, but not adult, mice leads to cognitive deficits

Abstract

Creatine (Cr) is a guanidino compound that provides readily available phosphate pools for the regeneration of spent adenosine triphosphate (ATP). The lack of brain Cr causes moderate to severe intellectual disability, language impairment, and epilepsy. The most prevalent cause of Cr deficiency are mutations in the X-linked SLC6A8 (Creatine transporter; CrT) gene, known as CrT deficiency (CTD). One of the most critical areas that need to be addressed is whether Cr is necessary for brain development. To address this concern, the Slc6a8 gene was knocked out in either neonatal (postnatal day (P)5) or adult (P60) mice using a tamoxifen-inducible Cre recombinase driven by the human ubiquitin C (UBC) promoter. Mice were tested in the Morris water maze, novel, object recognition, and conditioned fear 60 days after Slc6a8 deletion. In addition, overnight locomotor activity was analyzed. Mice that had the gene deleted on P5 showed deficits in the Morris water maze and novel object recognition, while there were no deficits in P60 knockout mice. Interestingly, the P5 knockout mice showed hyperactivity during the dark phase; however, when examining control mice, the effect was due to the administration of tamoxifen from P5 to 10. Taken together, the results of this study show that Cr is necessary during periods of brain development involved in spatial and object learning. This study also highlights the continued importance of using proper control groups for behavioral testing.

Figure 1.. Reductions in body weight and brain Cr levels following P5 or P60 Slc6a8 deletion.

A) Brain Cr levels in P5KO mice before and 60- or 90-days following tamoxifen exposure. By the time of behavioral testing (60 days) Cr levels were reduced to less than 10% of initial levels. B) Time course of Cr levels in P60KO mice. C) The P5KO mice gained weight as they aged however the growth did not match WT mice (main effect of group). D) In P60KO mice there was a body mass reduction following tamoxifen administration. Data are LSMEAN±SEM. N=13-18/group

Figure 2.. Spatial learning deficits in P5KO but not P60KO mice.

During visible platform training (A), P5KO mice show an increase latency on days 2 and 3 of training compared with WT-VEH mice. All mice improved across days. In the hidden platform training, the P5KO mice show higher mean latencies to the platform across all days of testing (Mean) during the (B) acquisition and (C) reversal phases of the MWM compared with the WT-VEH mice. (D) Shows the overall mean across days for all groups tested with only the P5KO mice differing from the WT-VEH mice. No differences were observed in the P60 during (E) visible platform, (F) acquisition, or (G) reversal testing. There were no differences between any P60 group tested and WT-VEH during the acquisition phase (H). For panels D and H, Slc6a8 refers to the Slc6a8 genotype: (+) is WT, (FL) is floxed; Cre refers to UBC-Cre: (−) means the mouse was not carrying a copy of UBC-Cre, (+) indicates expression; Tam refers to tamoxifen administration, (+) received tamoxifen, (−) received vehicle. The far-right bar in each panel shows the WT-VEH while the far-left bar of each panel shows the age-respective KO Data are LSMEAN±SEM, *p<0.05. N=13-18/group

Figure 3.. Novel object recognition deficits in P5KO mice.

Slc6a8 refers to the Slc6a8 genotype: (+) is WT, (FL) is floxed; Cre refers to UBC-Cre: (−) means the mouse was not carrying a copy of UBC-Cre, (+) indicates expression; Tam refers to tamoxifen administration, (+) received tamoxifen, (−) received vehicle. The far-right bar in each panel shows the WT-VEH while the far-left bar of each panel shows the age-respective KO. Left: In the P5-treated group, only the P5KO mice show a reduction in time spent with the novel object compared with the WT-VEH. Right: No differences from WT-VEH were observed in any of the groups treated at P60. Data are LSMEAN±SEM. N=9-18/group. *p<0.05

Figure 4.. Neonatal tamoxifen administration leads to hyperactivity during adulthood.

Locomotor activity was assessed in mice for 14 hours with the shaded region representing testing during the lights-off phase. Data are presented as the total beam breaks during a 20-min period. (A) The P5KO mice show a hyperactivity compared with WT-VEH mice; however, the remaining tamoxifen treated mice show a similar hyperactivity. (B) No differences in activity were observed between WT-VEH and P60K0. Data are LSMEAN±SEM. N=13-18/group. *p<0.05 P5KO vs WT-VEH; # p<0.05 Tamoxifen controls vs WT-VEH