FMRP(1–297)-tat restores ion channel and synaptic function in a model of Fragile X syndrome

Fragile X Syndrome results from a loss of Fragile X Mental Retardation Protein (FMRP). We now show that FMRP is a member of a Cav3-Kv4 ion channel complex that is known to regulate A-type

potassium current in cerebellar granule cells to produce mossy fiber LTP. Mossy fiber LTP is absent in Fmr1 knockout (KO) mice but is restored by FMRP(1-297)-tat peptide. This peptide

further rapidly permeates the blood-brain barrier to enter cells across the cerebellar-cortical axis that restores the balance of protein translation for at least 24 h and transiently

reduces elevated levels of activity of adult Fmr1 KO mice in the Open Field Test. These data reveal that FMRP(1-297)-tat can improve function from the levels of protein translation to

synaptic efficacy and behaviour in a model of Fragile X syndrome, identifying a potential therapeutic strategy for this genetic disorder.

The loss of Fragile X mental retardation protein (FMRP) in Fragile X syndrome (FXS) can interfere with the translation of numerous proteins important for the normal development of synaptic

transmission. A growing body of evidence suggests that behavioral disorders in FXS can involve a disruption in the processing of sensory information in both cortex and cerebellum1,2,3. The

mossy fiber projection to cerebellar granule cells represents a major synaptic portal to the cerebellar cortex where long-term plasticity can shape sensory input4,5,6,7. Any dysfunction at

this synapse will thus impair signal processing by the cerebellar cortex at the first stage of sensory input.

The present study shows that FMRP is a constituent member of the Cav3–Kv4 complex that modulates both Cav3 and Kv4 channels to reduce Kv4 current amplitude. LTP at the mossy fiber-granule

cell input is absent in Fmr1 knockout (KO) mice but rescued by infusing an N-terminal fragment of FMRP (FMRP(1–297)) into granule cells. Moreover, a FMRP(1–297)-tat peptide introduced to

Fmr1 KO mice by tail vein injection restores Cav3–Kv4 complex function and mossy fiber LTP, reduces the level of activity in adult animals within 1 h, and rescues disrupted translation of

select proteins associated with FXS for at least 24 h, supporting the potential for a tat-FMRP conjugate approach to be developed as a therapeutic agent for FXS.

The reduction in A-type current in granule cells following a theta burst stimulus (TBS) to mossy fibers was traced to a hyperpolarizing shift in the half voltage for Kv4 channel inactivation

(Vh) (referred to here as a left-shift in Kv4 Vh)8. To determine the potential role for FMRP in regulating Kv4 channels and LTP in granule cells, whole-cell recordings were obtained in the

vermis region of lobule 9 from male P16–P22 wild-type (WT) mice or Fmr1 KO mice and mossy fibers were stimulated to evoke a just threshold excitatory postsynaptic potential (EPSP) (Fig. 1a).

In 6/6 cells of WT mice a TBS was followed by an initial peak increase in EPSP amplitude that then decreased to an elevated level of 138.8 ± 11.0% (n = 6; p = 0.012) of control 5–13 min

post TBS (Fig. 1a). In WT cells just subthreshold synaptic stimulation was associated with little firing probability, but TBS invoked an increase in firing probability to single pulse

synaptic stimuli that persisted for at least 10 min post stimulation (Fig. 1a)8,15.

a–c Plots of the mean amplitude of the mossy fiber-evoked EPSP and probability of firing per stimulus in whole-cell recordings of lobule 9 granule cells. EPSP amplitudes were only calculated

for stimuli that were subthreshold to spike discharge and probability of spike firing was averaged for every 1-min interval (6 stimuli). a, b Theta burst stimulation (TBS, indicated by

arrow) of mossy fiber input evokes LTP of the EPSP and an increase in probability of firing in granule cells of WT mice (a) (% change of EPSP: 138.8 ± 11.0%; firing probability: resting

condition 0.5 ± 0.5%, after TBS 25.0 ± 11.4%, n = 6 cells from 6 mice) but not Fmr1 KO mice (b) (% change of EPSP: 100.6 ± 5.2%; firing probability: resting condition 1.4 ± 1.5%, after TBS

0.5 ± 0.6%, n = 7 cells from 7 mice). c Infusing 3 nM FMRP(1–297) into granule cells of Fmr1 KO mice rescues LTP of spike firing probability but not EPSP amplitude (% change of EPSP: 103.6 ±

 10.3%; firing probability: resting condition 9.5 ± 7.5%, after TBS 35.8 ± 10.4%, n = 7 cells from 6 mice). d–f Plots of the voltage for inactivation and activation of Kv4 current in granule

cells in resting conditions (control) and following TBS of mossy fiber input. Insets in (d–f) superimpose Kv4 current evoked by a step from −70 to −30 mV for either condition. Following TBS

Kv4 Vh and Va are left-shifted in WT mice (d) (Vh, p = 0.001; Va, p = 0.004) but not in Fmr1 KO mice (e) (Vh, p = 0.116; Va, p = 0.57). f Infusing 3 nM FMRP(1–297) into granule cells of

Fmr1 KO mice restores the ability for TBS stimulation to left-shift Kv4 Vh and Va to reduce Kv4 current amplitude within 10 min of introduction (Vh, p = 0.004; Va, p = 0.008). All mice were

P16–P22. Average values are mean ± s.e.m. All statistics were conducted with paired-sample Student’s t test. *p