The fragile X mental retardation protein in circadian rhythmicity and memory consolidation

Abstract

The control of new protein synthesis provides a means to locally regulate the availability of synaptic components necessary for dynamic neuronal processes. The fragile X mental retardation protein (FMRP), an RNA-binding translational regulator, is a key player mediating appropriate synaptic protein synthesis in response to neuronal activity levels. Loss of FMRP causes fragile X syndrome (FraX), the most commonly inherited form of mental retardation and autism spectrum disorders. FraX-associated translational dysregulation causes wide-ranging neurological deficits including severe impairments of biological rhythms, learning processes, and memory consolidation. Dysfunction in cytoskeletal regulation and synaptic scaffolding disrupts neuronal architecture and functional synaptic connectivity. The understanding of this devastating disease and the implementation of meaningful treatment strategies require a thorough exploration of the temporal and spatial requirements for FMRP in establishing and maintaining neural circuit function.

Figure 1. FMRP and the Cytoskeleton

The interactions between FMRP and both the microtubule and actin cytoskeleton elements are modeled. MAP1B, which impinges upon microtubule stability, is negatively regulated by FMRP (17, 83). This interaction normally provides for regulated synaptic deposition and growth. Regarding FMRP and the actin cytoskeleton, one known pathway involves the small GTPase Rac1 (92), which antagonizes the formation of the FMRP::CYFIP complex (93, 147). These components then negatively influence profilin functionality (89), which serves to promote actin filament polymerization. In the absence of FMRP, there is no attenuation of the resulting cellular protrusion causing neuronal overgrowth. FMRP also potentially alters actin dynamics by influencing the activation state of ADF/cofilin, which mediates filament severing and depolymerization, while also capable of nucleation. Though more often discussed in relation to AMPA receptor internalization in concert with FMRP and eEF2K (106), Arc is also linked to facilitating cofilin modifications that allow cytoskeletal expansion (245). As FMRP negatively regulates Arc translation, this could contribute to appropriate neuronal structuring. FMRP also negatively regulates the cofilin phosphatase, PP2A (96), which in turn can mediate dephosphorylation of FMRP creating a negative feedback loop (130, 131).

Figure 2. Drosophila Clock Circuitry

The Drosophila circadian clock circuitry is diagrammed with neuronal class positioning as indicated. Connectivity patterns are highlighted unilaterally with black arrows for clarity. Photic information is relayed from the compound eye (CE) via the Hofbauer-Buchner eyelet projecting to the central brain through the optic lobe (OL). Therein, the lateral neurons (LN) [dorso-LN (LN_d), the large and small ventro-LN (lLN_v and sLN_v, respectively), and the 5^th non-pigment dispersing factor expressing sLN_v] work in concert with the dorsal neurons (DN1, DN2, and DN3) to maintain rhythmicity. Other clock-associated neurons include the lateral posterior neurons (LPN), the crustacean cardioactive peptide expressing neurons (CCAP), and the prothoracic gland neurons (PG-LP). Lastly, the apparent target neurons of the pars intercerebralis in the dorsal protocerebrum (PI-3 neurons) receive input from several clock neuron classes. Modeled with permission after (170).

Figure 3. Null dfmr1 Abnormalities in Lateral Clock Neurons

A) Representative images of adult Drosophila central brain regions immunohistochemically labeled for pigment dispersing factor (PDF), a neuropeptide expressed only in ventrolateral neurons (LN_v). Arrow indicates the dorsal horn bifurcation point of the small ventrolateral neurons (sLN_v). The large ventrolateral neurons (lLN_v) mediate interhemisphere connectivity via the posterior optic tract (POT). Higher magnification views of the left and right sLN_v terminal axonal projections are shown. B) Cartoon summarizing the primary LN_v axonal phenotypes observed in null dfmr1 animals. The complex lLN_v dendritic and terminal axonal elaborations within each ipsilateral and contralateral optic lobe, respectively, have been omitted for simplicity. Wild-type LN_vs (left hemisphere) maintain stereotyped and contained projection pathway patterns. Null dfmr1 LN_vs (right hemisphere) display 1) overextension and mistargeting beyond their normal dorso-medial branching and defasiculation point, 2) an increased frequency of aberrant collateral branching, and 3) POT splitting (89, 137, 139, 147).

Figure 4. Drosophila Olfactory Learning and Memory Circuitry

Drosophila olfactory learning and memory involves a circuit that mediates the transduction of information initially encountered at the antennae via the olfactory receptor neurons (ORNs) to the Projection Neurons (PNs) organized within the glomeruli of the antennal lobe (AL). The PNs then extend to both the Mushroom Body Kenyon cells (MB-KC) and further to the lateral horn (LH). The synaptic field of the PN and MB-KC is spatially restricted to the region known as the calyx. Upon projection from the calyx and peduncle region (P), MB-KC bifurcation and projection patterns establish the MB lobes themselves (α and β indicated).

Figure 5. Null dfmr1 Abnormalities in Mushroom Body Neurons

A) The Drosophila Mushroom Body (MB) illuminated via the combination of a MB-specific GAL4 genetic element driving the expression of a UAS-GFP transgene. Note the bilateral symmetry of these structures. Anatomically, beyond the cell bodies (CB), individual MB Kenyon cells put forth their dendritic specializations, which, as stated, are concentrated in the calyx (cal). The axons then project together, as the fasiculated peduncle (P), before specific bifurcation and differential projection patterns then determine the MB neuron’s residence within a given axonal lobe (ax): αβ, α’β’, and γ. B) Cartoon summary of two well-characterized MB phenotypes associated with dfmr1 insufficiency. In dfmr1 null mutants, β-lobe axons fail to terminate approaching the midline, resulting in inappropriate extension into the contralateral MB lobe and β-lobe fusion (138, 143). At single cell resolution, null dfmr1 MB neurons display structural over-elaboration, including enhanced process initiation resulting in excessive overgrowth and branching of axonal processes (depicted) and increased dendritic field complexity (not shown) (15, 144, 146).