The FXG: a presynaptic fragile X granule expressed in a subset of developing brain circuits

Abstract

The loss of Fragile X mental retardation protein (FMRP) causes Fragile X syndrome, the most common inherited mental retardation and single gene cause of autism. Although postsynaptic functions for FMRP are well established, potential roles at the presynaptic apparatus remain largely unexplored. Here, we characterize the expression of FMRP and its homologs, FXR1P and FXR2P, in the developing, mature and regenerating rodent nervous system, with a focus on presynaptic expression. As expected, FMRP is expressed in the somatodendritic domain in virtually all neurons. However, FMRP is also localized in discrete granules (Fragile X granules; FXGs) in a subset of brain regions including frontal cortex, hippocampal area CA3 and olfactory bulb glomeruli. Immunoelectron microscopy shows that FMRP is localized at presynaptic terminals and in axons within these FXG-rich regions. With the exception of the olfactory bulb, FXGs are prominent only in the developing brain. Experiments in regenerating olfactory circuits indicate that peak FXG expression occurs 2-4 weeks after neurogenesis, a period that correlates with synapse formation and refinement. Virtually all FXGs contain FXR2P, while region-selective subsets harbor FMRP and/or FXR1P. Genetic studies show that FXR2P is essential for FXG expression, while FMRP regulates FXG number and developmental profile. These findings suggest that Fragile X proteins play a distinct, presynaptic role during discrete developmental epochs in defined circuits of the mammalian CNS. We propose that the neurological defects in Fragile X syndrome, including the autistic features, could be due in part to the loss of FMRP function in presynaptic compartments.

Figure 1.

Two classes of FMRP-containing structures in the olfactory bulb. A, FMRP was expressed in the somata and proximal dendrites of all major neuron types in the olfactory bulb. B, FMRP was expressed in olfactory sensory neuron somata in the olfactory epithelium (FMRP, green; OMP, red). C, High magnification of the glomerular layer revealed a second class of FMRP-containing structures, termed FXGs, that are discrete, granular, and selectively localized to the glomerular neuropil (arrows) and olfactory nerve layer (ONL) (arrowheads). Note that FXGs are not detected in the synapse-rich external plexiform layer (EPL; asterisk). Controls, Neither the somatodendritic staining nor FXGs were observed in experiments in which fmr1 null mice sections were incubated with anti-FMRP antibody (D), or wild-type sections were labeled with an equal concentration of irrelevant IgG (E). GL, Glomerular layer; MBL, mitral cell body layer; GCL, granule cell layer. Scale bars: A, 200 μm; B–E, 20 μm.

Figure 2.

FXGs are expressed presynaptically in olfactory glomeruli. FXGs, containing FMRP (green, A, B, D, E), were enriched within OSN axonal domains (arrows, A–F) marked by OMP (red, B, C). Conversely, relatively few FXGs were observed in dendrite-rich domains (arrowheads, A–F) marked by MAP2 (red, E and F). Immunogold electron microscopy showed that FMRP was localized in presynaptic specializations (arrow, G, H). Scale bars: A–F, 10 μm; G, H, 90 nm.

Figure 3.

FXGs are expressed in the hilus of the dentate gyrus and CA3 but not in CA1. FMRP was expressed in the cell bodies and proximal dendrites of nearly all P15 hippocampal neurons (A; epifluorescence microscopy). FXGs (arrows) were relatively sparse in the hilus of the dentate gyrus (B), with higher levels present in CA3 (C). Strikingly, no FXGs were observed in area CA1 (D). Boxes in A shown as confocal micrographs in B–D. Scale bars: A, 850 μm; B–D, 10 μm.

Figure 4.

Presynaptic and axonal FMRP in the hippocampus. Within CA3 stratum pyramidale of juvenile (P15) mice, FXGs were observed within the neuropil (arrows, A, B), specifically concentrating within the areas containing calbindin-positive (red, B) mossy fibers, the axonal projections from granule cells of the dentate gyrus. Immunogold electron microscopy of FMRP demonstrates its presence in the presynaptic apparatus (C) within mossy fiber terminals. D, Quantification of FMRP immunogold labeling in wild-type and fmr1 null mice. FMRP immunogold particles were more abundant at the mossy fiber presynaptic terminals (1.2 and 1.7 particles per 100 μm²) than in the postsynaptic apparatus (0.65 and 1.2 particles per 100 μm²). A knock-out mouse exhibited 0.101 presynaptic particles and 0.0504 postsynaptic per 100 μm² (see also supplemental Figs. S1, S2, S4, available at www.jneurosci.org as supplemental material). Scale bars: B (for A, B), 10 μm; C, 90 nm.

Figure 5.

FXG density is dynamically regulated during development. FXG expression was assessed at three developmental epochs: juvenile (P15), sexual maturation (P30) and adult (P60). In frontal cortex (A–C, D), motor cortex (D), visual cortex (D), and CA3 stratum oriens (E), a steady decline in granule density from juvenile to adult was observed. FXG density in frontal cortex declined from 0.25 ± 0.01 FXGs per 10 μm² (n = 13) at P15 to 0.119 ± 0.016 (n = 13) at P30 and 0.023 ± 0.004 (n = 11) at P60. Motor cortex density declined from 0.127 ± 0.010 (n = 6) at P15 to 0.023 ± 0.006 (n = 6) at P30. Visual cortex decreased from 0.148 ± 0.016 (n = 6) at P15 to 0.023 ± 0.010 (n = 5) at P30. Stratum oriens of CA3 declined from 0.642 ± 0.053 (n = 14) at P15, to 0.176 ± 0.021 (n = 12) at P30 and 0.045 ± 0.010 (n = 10) at P60. In contrast, FXG density in the olfactory bulb at P15 was 0.718 ± 0.063 (n = 39), followed by a peak of 1.52 ± 0.12 (n = 42) at P30, and subsequently a decline to 0.776 ± 0.081 (n = 38) at P60. Number of animals: 3 for P15 frontal cortex and hippocampus; 2 for all other points. Scale bar, 20 μm. Values expressed as mean ± SEM.

Figure 6.

FXG density during innervation and synaptogenesis in the regenerating olfactory system. Adult rats were treated with methyl bromide (MeBr), which ablates the olfactory epithelium and results in synchronized reinnervation and reformation of synaptic circuitry in the olfactory bulb (Schwob et al., 1995, 1999). FXG density was measured in olfactory glomeruli in untreated controls and at the indicated days post treatment (DPT). Control animals had 0.922 ± 0.075 FXGs per 10 μm² (n = 61). At 7 and 10 DPT, as axons in the glomeruli are dying, the numbers fell to 0.261 ± 0.061 (n = 21) and 0.151 ± 0.043 (n = 21), respectively. By 14 DPT, as axonal reinnervation of the glomeruli is beginning, levels rose to 0.352 ± 0.054 (n = 35). At 21 DPT, as glomerular reinnervation is largely complete, levels rose to 0.789 ± 0.100 (n = 37). By 28 DPT, as axons are integrating into the circuit, FXG density rose to 2.36 ± 0.20 (n = 41). Levels returned to baseline by 35 DPT (0.938 ± 0.127; n = 24). MeBr treatment had a significant effect on FXG density (p < 0.001, ANOVA). FXG density was significantly different from controls at 7, 10, 14, and 28 DPT (p < 0.01, Dunnett's t test). Number of animals: 4 for control, 14 DPT, 21 DPT, and 28 DPT; 2 for all other points. Values expressed as mean ± SEM.

Figure 7.

FXG composition in frontal cortex. Confocal micrographs of P15 frontal cortex. FXGs contain FMRP (green; A, B, D, E), FXR1P (red; B, C), and FXR2P (red; E, F). FXR1P, in addition to its somatic staining, colocalizes with FMRP in the FXGs (arrows, B) and is also present in additional granules that do not contain FMRP (arrowhead, B). FXR2P is expressed most strongly in FMRP-containing FXGs (arrows, E) with comparatively weak somatic staining. Scale bar, 10 μm.

Figure 8.

FXR2P and FMRP differentially regulate FXG expression. FXGs were detected in P30 wild-type frontal cortex as seen by FMRP (A, arrows) and FXR2P (B, arrows). In age-matched fxr2 knock-out mice, no FXGs were detected with anti-FMRP (C). In age-matched fmr1 knock-out mice, a significant increase in FXR2P-containing FXGs was observed (arrows, D). Quantitation of FXR2P granules in P15 and P30 frontal cortex (E) showed that FXG density in fmr1 KO mice at P15 was 1.03 ± 0.25 FXGs per 10 μm² (n = 9) compared with 0.253 ± 0.034 (n = 9) in wildtype. FXG density in fmr1 KO mice at P30 was 0.447 ± 0.036 (n = 9) compared with 0.127 ± 0.013 in wildtype (n = 9). Values expressed as mean ± SEM. A two-way ANOVA confirmed an effect for both age (p < 0.001) and genotype (p < 0.001), as well as an interaction between the two (p < 0.01). Number of animals: 4 for the P15 time points, 3 for the P30 time points. Scale bar, 10 μm.