Multiple Behavior Phenotypes of the Fragile-X Syndrome Mouse Model Respond to Chronic Inhibition of Phosphodiesterase-4D (PDE4D)

Abstract

Fragile-X syndrome (FXS) patients display intellectual disability and autism spectrum disorder due to silencing of the X-linked, fragile-X mental retardation-1 (FMR1) gene. Dysregulation of cAMP metabolism is a consistent finding in patients and in the mouse and fly FXS models. We therefore explored if BPN14770, a prototypic phosphodiesterase-4D negative allosteric modulator (PDE4D-NAM) in early human clinical trials, might provide therapeutic benefit in the mouse FXS model. Daily treatment of adult male fmr1 C57Bl6 knock-out mice with BPN14770 for 14 days reduced hyperarousal, improved social interaction, and improved natural behaviors such as nesting and marble burying as well as dendritic spine morphology. There was no decrement in behavioral scores in control C57Bl6 treated with BPN14770. The behavioral benefit of BPN14770 persisted two weeks after washout of the drug. Thus, BPN14770 may be useful for the treatment of fragile-X syndrome and other disorders with decreased cAMP signaling.

Conflict of interest statement

M.E.G is an employee of Tetra. Tetra owns BPN14770. All other authors declare that they have no competing interests.

Figures

Figure 1

Schematic Diagram Depicting Treatment Groups and Schedule of Events. Treatment groups were wild-type C57Bl6 mice exposed to vehicle (WT-Vehicle), wild-type C57Bl6 mice exposed to BPN14770 (WT-BPN14770), fmr1 KO mice exposed to vehicle, or fmr1 KO mice exposed to BPN14770. Three experiments were performed. In each experiment, mice were treated daily for 14 days by oral gavage with either vehicle or BPN14770. In the first experiment (Expt 1), mice were profiled for open field activity, social interaction, and natural behaviors (nesting & marble burying) on the 14th day of treatment. In the second experiment (Expt 2), mice were treated daily for 14 days, profiled for open field activity on the 14th day of treatment, then killed and prepared for spine morphometry. In the third experiment (Expt 3), mice were treated daily for 14 days, treatment was then stopped, and 14 days after cessation of treatment the mice were profiled for open field activity, social interaction, nesting & marble burying. Each treatment group contained 10 adult male mice.

Figure 2

BPN14770 Improved Behavioral Phenotypes of fmr1 KO Mice (A) Wild-type (WT) or fmr1 KO (KO) adult male mice were treated with vehicle or BPN14770 daily for 14 days (Expt 1) and then profiled for behavior. From left-to-right, data are shown for hyperarousal (Number of Squares crossed in the open field test), social interaction (Duration of Sniffing), nesting (Nesting Score), and marble burying (Marbles Buried). (B) Mice were treated daily for 14 days with vehicle or BPN14770 and then drug was withdrawn for 14 days (Expt 3). Mice were profiled for hyperarousal, social interaction, nesting and marble burying on the 14th day after drug withdrawal. Data were analyzed by Two-way ANOVA followed by Tukey’s post-test corrected for multiple comparisons. P-values shown are: n.s. not significant, *p < 0.05, **p < 0.01, ***p < 0.001 (N = 10 mice per group).

Figure 3

Representative Laser Scanning Confocal Micrographs Of Dendritic Segments And Dendritic Spines For Pyramidal Neurons In Medial Prefrontal Cortex. Dendritic spine morphometry was performed on dendritic segments of layer 3 and layer 5 pyramidal cells in medial prefrontal cortex (schematic). The sampling region for layer 3 pyramidal cells (L3P) was a segment of the primary apical dendrite extending from the soma to a distance of 100–150 µM. Layer 5 pyramidal cells were sampled from the soma to a distance of 200–250 µM of the primary apical dendrite and from the branch point of the apical tuft for a distance of 50 µM. High magnification insets for each segment sampled show the presence of thin-necked, immature spines (arrow) as well as mushroom (arrowhead only) and stubby spine (double arrowhead) types.

Figure 4

Dendritic Spine Multi-Dimensional Feature Distribution Plots. Multi-dimensional spine length x head diameter plots are shown for (A) the apical dendrite of Layer 3 pyramidal cells (N = 3,502 spines, 70 cells vehicle; N = 3,322 spines, 69 cells treated), (B) the apical dendrite of Layer 5 pyramidal cells (N = 3,792 spines, 70 cells vehicle; N = 3,443 spines, 69 cells treated), and (C) the apical dendrite tufts of Layer 5 pyramidal cells (N = 2,626 spines, 65 cells vehicle; N = 3,026 spines, 68 cells treated). The first two panels compare spines from fmr1 KO mice exposed to vehicle or BPN14770. The third panel shows the difference plot. The difference map (vehicle subtracted from BPN14770) shows maturation (shortening) of spines on the apical dendrites of layer 3 pyramidal neurons. There was no change in dendritic spine maturation on layer 5 pyramidal neurons.

Figure 5

Effect of BPN14770 on Dendritic Spine Length. The distribution of spine length is shown for the apical dendrite of Layer 3 pyramidal cells (L3P apical), the apical dendrite of Layer 5 pyramidal cells (L5P apical), and the apical dendrite tufts of Layer 5 pyramidal cells (L5P tufts). Dendritic spines were sampled for length, head dimeter and neck width. Data are plotted for vehicle (red circles) and BPN14770-treated (blue squares) fmr1 KO mice. Data were analyzed by the 2-sample Komolgorov-Smirnov test. P-values shown are ***p < 0.0001.