Paradoxical effect of baclofen on social behavior in the fragile X syndrome mouse model

Abstract

Introduction: Fragile X syndrome (FXS) is a common monogenetic cause of intellectual disability, autism spectrum features, and a broad range of other psychiatric and medical problems. FXS is caused by the lack of the fragile X mental retardation protein (FMRP), a translational regulator of specific mRNAs at the postsynaptic compartment. The absence of FMRP leads to aberrant synaptic plasticity, which is believed to be caused by an imbalance in excitatory and inhibitory network functioning of the synapse. Evidence from studies in mice demonstrates that GABA, the major inhibitory neurotransmitter in the brain, and its receptors, is involved in the pathogenesis of FXS. Moreover, several FXS phenotypes, including social behavior deficits, could be corrected in Fmr1 KO mice after acute treatment with GABA_B agonists.

Methods: As FXS would probably require a lifelong treatment, we investigated the effect of chronic treatment with the GABA_B agonist baclofen on social behavior in Fmr1 KO mice on two behavioral paradigms for social behavior: the automated tube test and the three-chamber sociability test.

Results: Unexpectedly, chronic baclofen treatment resulted in worsening of the FXS phenotypes in these behavior tests. Strikingly, baclofen treatment also affected wild-type animals in both behavioral tests, inducing a phenotype similar to that of untreated Fmr1 KO mice.

Conclusion: Altogether, the disappointing results of recent clinical trials with the R-baclofen enantiomer arbaclofen and our current results indicate that baclofen should be reconsidered and further evaluated before its application in targeted treatment for FXS.

Keywords: FMR1; baclofen; fragile X mental retardation protein; fragile X syndrome; targeted treatment; γ-aminobutyric acid.

Figure 1

Control experiment of matches between Fmr1 KO mice and WT littermates using aspartame drinking water, in the automated tube test. Fmr1 KO mice receiving aspartame drinking water win most of the matches from WT littermates receiving aspartame drinking water (p = .054 on day 1 to .003 on day 4, n = 4 mice per group). p values: ^{^}<.05, *<.01, **<.001

Figure 2

The effect of 1 mg/ml baclofen treatment on the behavior of Fmr1 KO mice and wild‐type littermates in the automated tube test and the three‐chamber sociability test. (a–c) Results of the automated tube test, expressed in percentage of matches won. (a) Fmr1 KO mice treated with baclofen win most matches against WT littermates receiving aspartame drinking water, winning 80% of the matches on day 1, which increases to over 90% on day 4 (binomial test, p = .000, n = 6 per group). (b) Fmr1 KO mice treated with 1 mg/ml baclofen water win most matches against Fmr1 KO mice receiving aspartame water, winning more than 80% of the matches (p = .000, n = 6 per group). (c) WT littermates treated with 1 mg/ml baclofen win most matches against WT littermates receiving aspartame water, winning more than 90% of the matches (p = .000, n = 6 per group). (d) Sociability behavior, in the three‐chamber sociability test, expressed as average time spent sniffing the stranger mouse, compared to the time spent sniffing the empty cage. All groups sniff the stranger mouse more than the empty cage (two‐way RM ANOVA with within‐subject factor side of stranger, p = .000). Fmr1 KO mice receiving aspartame water sniff the stranger mouse significantly more than WT littermates receiving aspartame water (two‐way RM ANOVA with stranger side as within‐subject factor and genotype as between‐subject factor, p < .003). WT littermates receiving baclofen water sniff the stranger mouse more than WT littermates receiving aspartame water (p = .008), making baclofen‐treated WT littermates comparable to Fmr1 KO mice receiving aspartame or baclofen. The average time sniffing the stranger was similar for Fmr1 KO mice with and without baclofen treatment (RM ANOVA stranger side * treatment, p = .869). Data are expressed as mean ± SE; WT aspartame n = 13, Fmr1 KO aspartame n = 13, WT baclofen n = 11, Fmr1 KO baclofen n = 12. (e) Locomotor activity during sociability phase of the three‐chamber sociability test. WT and Fmr1 KO receiving baclofen drinking water show increased locomotor activity compared to WT and Fmr1 KO mice receiving aspartame water (two‐sided independent‐samples t‐test, p = .001 and p = .034, respectively). Data presented as mean ± SE. p‐Values: ^{^}<.05, *<.01, **<.001

Figure 3

The effect of 0.25 mg/ml baclofen treatment on the behavior of Fmr1 KO mice and WT littermates in the automated tube test and the three‐chamber sociability test. (a–c) Results of the automated tube test expressed in percentage of matches won. (a) Fmr1 KO mice treated with 0.25 mg/ml baclofen win most matches against WT littermates receiving aspartame drinking water (binomial test, 70%–90%, p = .000, n = 6 per group). (b) Fmr1 KO mice treated with 0.25 mg/ml baclofen water win most matches from Fmr1 KO mice receiving aspartame water, winning 75% of the matches on day 1, which increases to 100% on day 4 (p = .002 on day 1 to p = .000 on day 4, n = 6 per group). (c) WT littermates treated with 0.25 mg/ml baclofen win most matches from WT littermates receiving aspartame water, winning 80%–100% of matches (p = .01 on day 1 to p = .000 on day 4, n = 4 per group). (d) Three‐chamber sociability test. Sociability behavior, expressed as average time spent sniffing the stranger mouse, compared to the time spent sniffing the empty cage. WT aspartame n = 10, Fmr1 KO aspartame n = 10, WT baclofen n = 10, Fmr1 KO baclofen n = 10. All groups sniff the stranger mouse more than the empty cage (two‐way RM ANOVA with within‐subject factor side of stranger, p = .000). Fmr1 KO mice receiving aspartame water sniff the stranger mouse significantly more than WT littermates receiving aspartame water (two‐way RM ANOVA, p = .004). WT littermates receiving baclofen water sniff the stranger mouse more than WT littermates receiving aspartame water (two‐way RM ANOVA, p = .011), making baclofen‐treated WT mice comparable to Fmr1 KO mice receiving aspartame or Fmr1 KO mice receiving baclofen. Data are expressed as mean ± SE. (e) Locomotor activity during sociability phase of the three‐chamber sociability test. The locomotor activity does not differ between treatments. Fmr1 KO mice receiving baclofen have increased locomotor activity compared to WT littermates receiving baclofen (two‐sided independent‐samples t‐test, p = .000). Data presented as mean ± SE. p‐Values: ^{^}<.05, *<.01, **<.001