TY - JOUR
AU - Fallon, Justin R.
AB - Abstract Local mRNA translation in growing axons allows for rapid and precise regulation of protein expression in response to extrinsic stimuli. However, the role of local translation in mature CNS axons is unknown. Such a mechanism requires the presence of translational machinery and associated mRNAs in circuit-integrated brain axons. Here we use a combination of genetic, quantitative imaging and super-resolution microscopy approaches to show that mature axons in the mammalian brain contain ribosomes, the translational regulator FMRP and a subset of FMRP mRNA targets. This axonal translational machinery is associated with Fragile X granules (FXGs), which are restricted to axons in a stereotyped subset of brain circuits. FXGs and associated axonal translational machinery are present in hippocampus in humans as old as 57 years. This FXG-associated axonal translational machinery is present in adult rats, even when adult neurogenesis is blocked. In contrast, in mouse this machinery is only observed in juvenile hippocampal axons. This differential developmental expression was specific to the hippocampus, as both mice and rats exhibit FXGs in mature axons in the adult olfactory system. Experiments in Fmr1 null mice show that FMRP regulates axonal protein expression but is not required for axonal transport of ribosomes or its target mRNAs. Axonal translational machinery is thus a feature of adult CNS neurons. Regulation of this machinery by FMRP could support complex behaviours in humans throughout life. Introduction Neurons are exquisitely compartmentalized such that they can convey, process and store information at vast distances from the cell soma. A striking example of such topographic specialization is the axonal arbour, which is typically complex and often conveys information to widely distributed brain regions. Individual axonal branches within a single arbour can exhibit distinct morphologies, synaptic release properties and activity-dependent structural and functional regulation (1). A fundamental question is how such morphological and functional diversity is achieved, modified and maintained within a complex axonal arbour controlled by a single soma. This problem is particularly acute in the adult human nervous system where axons can extend tens of centimetres. One conceptually attractive solution to this challenge is the local translation of mRNAs in the axon. Translation in developing and regenerating axons is well documented (2,3), and recent evidence suggests that CNS axons in the mammalian brain may also utilize this mechanism (4,5). However the extent to which endogenous ribosomes, mRNA and translational regulators localize to adult brain axons that are integrated into mature circuitry is unknown (6,7). FMRP (Fragile X mental retardation protein) is an attractive candidate for regulating local axonal translation. FMRP is an RNA binding protein that serves as a critical regulator of experience-dependent translation in neurons (8). Loss of FMRP perturbs experience-dependent plasticity and results in the autism-related disorder Fragile X syndrome (FXS) (8–,13). In addition to its role in the somatodendritic compartment, FMRP also associates with a distinct granule termed the FXG (Fragile X granule). In contrast to the ubiquitous expression of FMRP in the somatodendritic domains of immature and adult neurons, FXGs are restricted to axonal tracts and neuropil in stereotyped brain circuits. In mouse brain, FXGs show striking temporal regulation that correlates with epochs of robust synaptic plasticity (14,15). Several lines of evidence have established that FXGs are exclusively axonal: 1) colocalization with axonal but not somatodendritic markers; 2) ultrastructural demonstration of FMRP localization in axons and presynaptic boutons; 3) identification of FXG-containing axonal tracts within specific circuits in the intact brain; 4) loss of FXGs in axonal tracts and target neuropil following chemical ablation of projection neurons; and 5) selective depletion of FMRP from FXGs following circuit-specific genetic ablation of FMRP in projection neurons (14,15). As an endogenous, readily identifiable and uniquely axonal structure containing RNA binding proteins, the FXG is an attractive portal for investigating translational machinery in axons in the brain. Here we used confocal and super resolution microscopy to identify FXGs in multiple brain regions and ask whether these axonal structures associate with ribosomes and identified mRNAs. This approach overcomes a key challenge to the detection of axonal translational machinery using electron microscopy, which is restricted to analysis of small volumes of individual brain regions. We used this approach to investigate the localization, composition and regulation of FXG-associated translational machinery in mouse, rat and human CNS axons. FXGs associate with ribosomes and a subset of FMRP-target mRNAs including transcripts encoding β-catenin and OMP (olfactory marker protein), proteins known to play a role in neuronal plasticity. Analysis of Fmr1 null mice provides evidence that FMRP regulates axonal translation but is not required for axonal transport of its target mRNAs. Surprisingly, hippocampal FXGs are prominent in adult rats but are restricted to juvenile stages in mice. Finally, axons in the adult human hippocampus contain essential components for protein synthesis including FXG-associated ribosomes and mRNA. Axons in the mammalian nervous system therefore contain ribosomes, mRNA and translational regulators that are positioned to contribute to neuronal maintenance and plasticity throughout life. Results Quantification of Fragile X granule components FXGs always contain FXR2P, exhibit a characteristic morphology, are present in fibre tracts and neuropil and are restricted to specific circuits (c.f., Fig. 1). In previous work we took advantage of these traits to develop standardized image analyses for identifying FXGs. FXGs identified using these approaches always colocalized with axonal markers in the brain and were not observed to colocalize with markers of neuronal cell bodies or dendrites (14,15). Combining these approaches with genetic and ablation experiments established that FXGs are restricted to axons and are not present in glia or in neuronal somata or dendrites (14,15). Figure 1. Open in new tabDownload slide FXG identification in olfactory sensory neurons. (A) Cre-dependent tdTomato reporter expression in an olfactory bulb from a #123-Cre;R26tdT/WT mouse expressing Cre under control of the olfactory sensory neuron-specific #123 promoter. tdTomato expression (red) was observed in olfactory sensory neuron axons but not in cells resident within the olfactory bulb (DAPI; blue). (B,C) Olfactory glomerulus (representative of boxed region in A) from a #123-Cre;R26tdT/WT mouse. tdTomato (red) in olfactory sensory neuron axons colocalizes with FXR2P (green) in FXGs (arrows) but not with FXR2P in cell bodies of resident olfactory bulb neurons (arrowheads). (D–M) Analysis of FMRP signal in FXGs and resident olfactory bulb neurons in wild type (#123-Cre;Fmr1wt/y) and cKO (#123-Cre;Fmr1flox/y) mice in which a floxed Fmr1 allele was selectively ablated in olfactory sensory neurons. (D,G) Schematic of FMRP and FXG expression in experiments depicted in E-F, H-M. Images were collected in olfactory bulb glomeruli. Image processing of the FXR2P signal was used to identify FXGs in these images. (E,F,H,I) Representative images of data analysed in J-M. Image processing of FXR2P immunostaining (red) identifies FXGs (blue). (E,F) In wild type brains, FMRP (green) is present in FXGs in OSN axons as well as in cell bodies of olfactory bulb neurons. (H-I) In cKO brains, FMRP is selectively ablated from FXGs in OSN axons but retained in cell bodies of olfactory bulb neurons. (J) Quantitation of FMRP levels in FXGs. FMRP intensity was quantified for each FXG. FMRP intensity in olfactory FXGs was significantly different between cKO (0.93 ± 0.012) and wild type brains (2.84 ± 0.022; P = 0.003661 by ANCOVA). Background fluorescence: 1.00 ± 0.0078. (K) Cumulative frequency distribution of average FMRP fluorescence intensity in olfactory FXGs for each genotype. Dashed line indicates the median signal in wild type FXGs. (L) Quantitation of FMRP levels in olfactory bulb somata. For each image, the average somatic signal in cells was calculated. There was no change in FMRP intensity in these cells between wild type (2.47 ± 0.027) and cKO (2.35 ± 0.019) animals (P = 0.7437 by ANCOVA). Background fluorescence: 0.95 ± 0.011. (M) Cumulative frequency distribution of average FMRP fluorescence intensity in olfactory bulb neurons. Dashed line indicates the median signal for wild type somata. For J-M, FMRP fluorescence in wild type and cKO regions of interest was compared using background fluorescence as a covariate. Background is depicted separately to provide context for interpreting FMRP signal. For all panels, arrows: FXGs; arrowheads: cell bodies. OE: olfactory epithelium; OB: olfactory bulb; tdT: tdTomato; OSN: olfactory sensory neuron. Scale bar = 500 μm in A; 20 μm in B; 10 μm in C; 15 μm in E,G; 5 μm in F,H. Wild type: n = 2720 FXGs and 325 cell bodies; 9 images; 3 animals. cKO: n = 2146 FXGs and 461 cell bodies; 9 images; 2 animals. Figure 1. Open in new tabDownload slide FXG identification in olfactory sensory neurons. (A) Cre-dependent tdTomato reporter expression in an olfactory bulb from a #123-Cre;R26tdT/WT mouse expressing Cre under control of the olfactory sensory neuron-specific #123 promoter. tdTomato expression (red) was observed in olfactory sensory neuron axons but not in cells resident within the olfactory bulb (DAPI; blue). (B,C) Olfactory glomerulus (representative of boxed region in A) from a #123-Cre;R26tdT/WT mouse. tdTomato (red) in olfactory sensory neuron axons colocalizes with FXR2P (green) in FXGs (arrows) but not with FXR2P in cell bodies of resident olfactory bulb neurons (arrowheads). (D–M) Analysis of FMRP signal in FXGs and resident olfactory bulb neurons in wild type (#123-Cre;Fmr1wt/y) and cKO (#123-Cre;Fmr1flox/y) mice in which a floxed Fmr1 allele was selectively ablated in olfactory sensory neurons. (D,G) Schematic of FMRP and FXG expression in experiments depicted in E-F, H-M. Images were collected in olfactory bulb glomeruli. Image processing of the FXR2P signal was used to identify FXGs in these images. (E,F,H,I) Representative images of data analysed in J-M. Image processing of FXR2P immunostaining (red) identifies FXGs (blue). (E,F) In wild type brains, FMRP (green) is present in FXGs in OSN axons as well as in cell bodies of olfactory bulb neurons. (H-I) In cKO brains, FMRP is selectively ablated from FXGs in OSN axons but retained in cell bodies of olfactory bulb neurons. (J) Quantitation of FMRP levels in FXGs. FMRP intensity was quantified for each FXG. FMRP intensity in olfactory FXGs was significantly different between cKO (0.93 ± 0.012) and wild type brains (2.84 ± 0.022; P = 0.003661 by ANCOVA). Background fluorescence: 1.00 ± 0.0078. (K) Cumulative frequency distribution of average FMRP fluorescence intensity in olfactory FXGs for each genotype. Dashed line indicates the median signal in wild type FXGs. (L) Quantitation of FMRP levels in olfactory bulb somata. For each image, the average somatic signal in cells was calculated. There was no change in FMRP intensity in these cells between wild type (2.47 ± 0.027) and cKO (2.35 ± 0.019) animals (P = 0.7437 by ANCOVA). Background fluorescence: 0.95 ± 0.011. (M) Cumulative frequency distribution of average FMRP fluorescence intensity in olfactory bulb neurons. Dashed line indicates the median signal for wild type somata. For J-M, FMRP fluorescence in wild type and cKO regions of interest was compared using background fluorescence as a covariate. Background is depicted separately to provide context for interpreting FMRP signal. For all panels, arrows: FXGs; arrowheads: cell bodies. OE: olfactory epithelium; OB: olfactory bulb; tdT: tdTomato; OSN: olfactory sensory neuron. Scale bar = 500 μm in A; 20 μm in B; 10 μm in C; 15 μm in E,G; 5 μm in F,H. Wild type: n = 2720 FXGs and 325 cell bodies; 9 images; 3 animals. cKO: n = 2146 FXGs and 461 cell bodies; 9 images; 2 animals. In the current study, we sought to determine whether translational machinery is associated with FXGs in axons in the intact brain. To achieve this goal, we developed methods to distinguish the translational machinery associated with FXGs from that localized in the somatodendritic domain or in non-neuronal cells. To test this approach, we applied it to the detection of FMRP, which is a component of the FXG and is also present in dendrites, neuronal cell bodies and glial cells. We reasoned that selectively ablating FMRP from projection neurons would result in loss of FXG-associated FMRP in axons innervating the target region, but that FMRP expression in the resident target neurons and associated non-neuronal cells would be unaffected. The design of these experiments, which were performed in two brain regions, is shown in Figs. 1D, G and Supplementary Material, S1A. We first tested this approach in the olfactory system, where FXGs are found in olfactory sensory neuron axons innervating olfactory glomeruli in the olfactory bulb (14,15). For these experiments, we used mice in which the #123 promoter drives Cre recombinase in olfactory sensory neurons but not in the resident cells of the olfactory bulb (16–18). The specificity of this driver was confirmed by crossing mice expressing #123-Cre with a strain harbouring the tdTomato reporter (R26loxP-STOP-loxP-tdTomato) (19). As shown in Fig. 1A, tdTomato fluorescence was observed in the olfactory sensory axons but was not detected in their target olfactory bulb neurons. Moreover, as expected, FXGs were only detected in tdTomato-expressing olfactory sensory neuron axons (Fig. 1B and C). We then asked whether crossing the #123-Cre mice to animals carrying a Cre-sensitive FMRP allele (20) resulted in selective depletion of FMRP from the FXGs in olfactory sensory neuron axons (Fig. 1D–M). We collected images from olfactory glomeruli immunolabelled for both FXR2P and FMRP. We performed an image analysis of the FXR2P signal to create regions of interest corresponding to either FXGs or cell bodies (see Methods). We then measured the FMRP signal in both the FXG and the cell body regions of interest. As expected, in wild type mice FMRP was found in FXGs as well as in cell bodies of olfactory bulb neurons (Fig. 1E and F, J–M). In contrast, no FXG-associated FMRP was detected in olfactory sensory neuron axons from #123-Cre:FMRPloxP mice (Fig. 1H–K). As predicted, FMRP was readily observed in cell bodies and dendrites of olfactory bulb neurons (Fig. 1H and I, L and M). We also tested this approach for selectively detecting FXG-associated proteins in the thalamus, where FXGs are found in axons of ventral anterior and ventrolateral (VAL) thalamic nuclei that project to the thalamic reticular nucleus (TRN) (14). For these experiments, we used mice in which a Syn1-Cre promoter drives Cre in restricted neuronal populations (21). In previous work, we confirmed the restricted expression of this transgene by crossing these mice to conditional FMRP knockout mice and showing that Cre-mediated recombination occurs in neurons in thalamic relay nuclei including VAL but not in neurons in the TRN (14). Supplementary Material, Fig. S1 shows that Synapsin-Cre-mediated conditional knockout of FMRP resulted in loss of FMRP from FXGs in VAL axons innervating the TRN, but not from cells resident within the TRN. Taken together with the olfactory bulb experiments described above, these findings provide additional evidence that FXGs are localized exclusively in axons and establish that this approach can accurately and quantitatively identify FXG-associated components. Ribosomes and mRNA in brain axons We then used this approach to investigate whether axonal translational machinery associates with FXGs. We first examined juvenile (P15) mouse brain, the age where FXGs are most broadly expressed in mice (15). To test whether ribosomes are co-localized with FXGs we used multiple distinct probes to visualize components of both the large (5S/5.8S and 28S rRNA) and small (18S rRNA and S6 protein) ribosomal subunits (Table 1). Using standardized image analyses (as above), we observed all of these ribosomal components in FXGs in axons of multiple brain regions including the olfactory bulb, motor cortex and hippocampus (Fig. 2, Supplementary Material, Fig. S2). The specificity of the detection of 5S/5.8S rRNA by the Y10b antibody was validated by confirming that the signal was reduced in RNase-treated tissue (Supplementary Material, Fig. S2). Interestingly, not all FXGs contained detectable ribosomes (conservatively defined as colocalizing with the top 1% of the ribosome signal within the image) as we identified rRNA in approximately one half of FXGs and S6 in approximately one sixth of FXGs (Table 2). To test the possibility that the observed colocalization reflects ribosomes in adjacent processes rather than direct association with FXGs, we examined granule composition and structure using structured illumination super-resolution microscopy of CLARITY-processed sections. We identified FXGs that visually contained Y10b immunofluorescence and asked whether this represented overlapping signal in the super-resolution reconstructed images. Of 24 such FXGs, all 24 exhibited direct association of FXGs with ribosomes (Fig. 2D). This analysis also revealed that FMRP, FXR2P and ribosomes are each enriched in overlapping yet distinct FXG subdomains and established an upper boundary of FXG diameter as ∼200 nm (Fig. 2D, Supplementary Material, Fig. S2D and E). Thus, axons in the brain contain ribosomes that associate with discrete granules, the FXGs. Figure 2. Open in new tabDownload slide FXGs associate with ribosomes in the juvenile mouse. (A) Schematic indicating location of depicted FXGs. Confocal micrographs from P15 mouse brains show that FXGs associate with: (B-E) 5S/5.8S rRNA including in neurofilament-expressing axons; (F) 18S and 28S rRNA; and (G) ribosomal protein S6. Quantification of this colocalization is in Table 2. For all panels, arrows: FXGs; arrowheads: cell bodies. SL: stratum lucidum; SP: stratum pyramidale. Scale bar = 10 μm in B; 5 μm in C, E-G; 450 nm in D. Figure 2. Open in new tabDownload slide FXGs associate with ribosomes in the juvenile mouse. (A) Schematic indicating location of depicted FXGs. Confocal micrographs from P15 mouse brains show that FXGs associate with: (B-E) 5S/5.8S rRNA including in neurofilament-expressing axons; (F) 18S and 28S rRNA; and (G) ribosomal protein S6. Quantification of this colocalization is in Table 2. For all panels, arrows: FXGs; arrowheads: cell bodies. SL: stratum lucidum; SP: stratum pyramidale. Scale bar = 10 μm in B; 5 μm in C, E-G; 450 nm in D. Table 1. Association of ribosomes and mRNAs with FXGs Components of the large and small ribosomal subunits as well as mRNAs associate with FXGs Potential FXG Component . Common Name . FXG Association . Somatic Expression . Large ribosomal subunit 5S/5.8S rRNA + + 28S rRNA + + Small ribosomal subunit 18S rRNA + + S6 ribosomal protein + + FXG-associated mRNAs Ctnnb1 β-catenin + + Map1b MAP1b + + Ncam1 NCAM + + Omp OMP + + mRNAs not detected in FXGs Bsn Bassoon – + Calm1 Calmodulin 1 – + Calm3 Calmodulin 3 – + Camk2b CaM Kinase II β – + Camk2n1 CaM Kinase II Inhibitor 1 – + Dlg4 PSD-95 – + Fmr1 FMRP – + Fxr1 FXR1P – + Fxr2 FXR2P – + Nrxn1 Neurexin 1 – + Nrxn2 Neurexin 2 – + Nrxn3 Neurexin 3 – + Olfm1 Olfactomedin 1 – + Pclo Piccolo – + Ppp3ca Calcineurin A – + Snap25 SNAP-25 – + Tcf4 TCF4 – + Potential FXG Component . Common Name . FXG Association . Somatic Expression . Large ribosomal subunit 5S/5.8S rRNA + + 28S rRNA + + Small ribosomal subunit 18S rRNA + + S6 ribosomal protein + + FXG-associated mRNAs Ctnnb1 β-catenin + + Map1b MAP1b + + Ncam1 NCAM + + Omp OMP + + mRNAs not detected in FXGs Bsn Bassoon – + Calm1 Calmodulin 1 – + Calm3 Calmodulin 3 – + Camk2b CaM Kinase II β – + Camk2n1 CaM Kinase II Inhibitor 1 – + Dlg4 PSD-95 – + Fmr1 FMRP – + Fxr1 FXR1P – + Fxr2 FXR2P – + Nrxn1 Neurexin 1 – + Nrxn2 Neurexin 2 – + Nrxn3 Neurexin 3 – + Olfm1 Olfactomedin 1 – + Pclo Piccolo – + Ppp3ca Calcineurin A – + Snap25 SNAP-25 – + Tcf4 TCF4 – + Open in new tab Table 1. Association of ribosomes and mRNAs with FXGs Components of the large and small ribosomal subunits as well as mRNAs associate with FXGs Potential FXG Component . Common Name . FXG Association . Somatic Expression . Large ribosomal subunit 5S/5.8S rRNA + + 28S rRNA + + Small ribosomal subunit 18S rRNA + + S6 ribosomal protein + + FXG-associated mRNAs Ctnnb1 β-catenin + + Map1b MAP1b + + Ncam1 NCAM + + Omp OMP + + mRNAs not detected in FXGs Bsn Bassoon – + Calm1 Calmodulin 1 – + Calm3 Calmodulin 3 – + Camk2b CaM Kinase II β – + Camk2n1 CaM Kinase II Inhibitor 1 – + Dlg4 PSD-95 – + Fmr1 FMRP – + Fxr1 FXR1P – + Fxr2 FXR2P – + Nrxn1 Neurexin 1 – + Nrxn2 Neurexin 2 – + Nrxn3 Neurexin 3 – + Olfm1 Olfactomedin 1 – + Pclo Piccolo – + Ppp3ca Calcineurin A – + Snap25 SNAP-25 – + Tcf4 TCF4 – + Potential FXG Component . Common Name . FXG Association . Somatic Expression . Large ribosomal subunit 5S/5.8S rRNA + + 28S rRNA + + Small ribosomal subunit 18S rRNA + + S6 ribosomal protein + + FXG-associated mRNAs Ctnnb1 β-catenin + + Map1b MAP1b + + Ncam1 NCAM + + Omp OMP + + mRNAs not detected in FXGs Bsn Bassoon – + Calm1 Calmodulin 1 – + Calm3 Calmodulin 3 – + Camk2b CaM Kinase II β – + Camk2n1 CaM Kinase II Inhibitor 1 – + Dlg4 PSD-95 – + Fmr1 FMRP – + Fxr1 FXR1P – + Fxr2 FXR2P – + Nrxn1 Neurexin 1 – + Nrxn2 Neurexin 2 – + Nrxn3 Neurexin 3 – + Olfm1 Olfactomedin 1 – + Pclo Piccolo – + Ppp3ca Calcineurin A – + Snap25 SNAP-25 – + Tcf4 TCF4 – + Open in new tab Table 2. Association of ribosomes, mRNA, and FXGs in wild type and Fmr1 null mouse brains. A significant fraction of FXGs associate with ribosomes and RNA. However, we note that the exact percentages of ribosome- and RNA-associated FXGs that we observed likely represent lower bounds, as some FXGs might associate with ribosomes and RNA below our limits of detection. All observations are significantly higher than would be expected by chance (P < 0.0001 by χ2 test; as compared to an expected colocalization of 5% for 5S/5.8S, polyA+ and rpS6 or an expected colocalization of 0.25% for Ctnnb1 and Omp mRNAs). † = FXG association with ribosomal RNA (P < 0.0001, two-tailed Fisher’s exact test) and Omp mRNA colocalization with FXGs (P = 0.0278, two-tailed Fisher’s exact test) were significantly different between wild type and Fmr1 null animals. No other component exhibited differential FXG association between wild type and Fmr1 null (P > 0.05, two-tailed Fisher’s exact test). n indicates the number of animals. n.d. = not determined. . . Wild type . Fmr1 null . FXG colocalization with: 5S/5.8S rRNA (Y10b) 52.4% (972/1856; n = 5) 34% (68/200; n = 2)† polyA+ RNA 33.3% (156/468; n = 5) 33.2% (137/413; n = 3) rpS6 14.3% (173/1214; n = 2) n.d. Ctnnb1 mRNA 24.8% (54/218; n = 3) 29.2% (71/243; n = 2) Omp mRNA 12.7% (53/417; n = 4) 8.51% (20/235; n = 3) Omp mRNA colocalization with: FXGs 16.7% (53/317; n = 4) 9.7% (20/206; n = 3)† . . Wild type . Fmr1 null . FXG colocalization with: 5S/5.8S rRNA (Y10b) 52.4% (972/1856; n = 5) 34% (68/200; n = 2)† polyA+ RNA 33.3% (156/468; n = 5) 33.2% (137/413; n = 3) rpS6 14.3% (173/1214; n = 2) n.d. Ctnnb1 mRNA 24.8% (54/218; n = 3) 29.2% (71/243; n = 2) Omp mRNA 12.7% (53/417; n = 4) 8.51% (20/235; n = 3) Omp mRNA colocalization with: FXGs 16.7% (53/317; n = 4) 9.7% (20/206; n = 3)† Open in new tab Table 2. Association of ribosomes, mRNA, and FXGs in wild type and Fmr1 null mouse brains. A significant fraction of FXGs associate with ribosomes and RNA. However, we note that the exact percentages of ribosome- and RNA-associated FXGs that we observed likely represent lower bounds, as some FXGs might associate with ribosomes and RNA below our limits of detection. All observations are significantly higher than would be expected by chance (P < 0.0001 by χ2 test; as compared to an expected colocalization of 5% for 5S/5.8S, polyA+ and rpS6 or an expected colocalization of 0.25% for Ctnnb1 and Omp mRNAs). † = FXG association with ribosomal RNA (P < 0.0001, two-tailed Fisher’s exact test) and Omp mRNA colocalization with FXGs (P = 0.0278, two-tailed Fisher’s exact test) were significantly different between wild type and Fmr1 null animals. No other component exhibited differential FXG association between wild type and Fmr1 null (P > 0.05, two-tailed Fisher’s exact test). n indicates the number of animals. n.d. = not determined. . . Wild type . Fmr1 null . FXG colocalization with: 5S/5.8S rRNA (Y10b) 52.4% (972/1856; n = 5) 34% (68/200; n = 2)† polyA+ RNA 33.3% (156/468; n = 5) 33.2% (137/413; n = 3) rpS6 14.3% (173/1214; n = 2) n.d. Ctnnb1 mRNA 24.8% (54/218; n = 3) 29.2% (71/243; n = 2) Omp mRNA 12.7% (53/417; n = 4) 8.51% (20/235; n = 3) Omp mRNA colocalization with: FXGs 16.7% (53/317; n = 4) 9.7% (20/206; n = 3)† . . Wild type . Fmr1 null . FXG colocalization with: 5S/5.8S rRNA (Y10b) 52.4% (972/1856; n = 5) 34% (68/200; n = 2)† polyA+ RNA 33.3% (156/468; n = 5) 33.2% (137/413; n = 3) rpS6 14.3% (173/1214; n = 2) n.d. Ctnnb1 mRNA 24.8% (54/218; n = 3) 29.2% (71/243; n = 2) Omp mRNA 12.7% (53/417; n = 4) 8.51% (20/235; n = 3) Omp mRNA colocalization with: FXGs 16.7% (53/317; n = 4) 9.7% (20/206; n = 3)† Open in new tab We next asked whether mRNA is also associated with these axonal granules. In situ hybridization revealed that FXGs associate with polyadenylated (polyA+) RNA (Fig. 3B, Supplementary Material, Fig. S3B). As with the ribosomes, the top 1% of this RNA signal was detected in a subset of FXGs (Table 2). To identify specific mRNAs associated with these granules, we performed in situ hybridization for a panel of 21 candidates, including those that were both FMRP targets and reported to be present in axons of cultured neurons (Table 1, S1)(6,22,23). Most striking was association of FXGs with Ctnnb1 (β-catenin) mRNA (Fig. 3C and D, Supplementary Material, Fig. S3H and I). This transcript associated with a large fraction of neocortical and hippocampal FXGs (Table 2). As expected from the results presented above, Ctnnb1 transcripts and ribosomes colocalized within FXGs (Fig. 3D, Supplementary Material, Fig. S3I). We observed several additional mRNAs associated with FXGs, including Map1b and Ncam1 (Supplementary Material, Fig. S3F and G; Table 1). Finally, we considered an additional candidate transcript, Omp (olfactory marker protein), since the Omp mRNA and FXGs are both expressed within axons of olfactory sensory neurons (14,15,24,25). As shown in Fig. 3E and K, Omp mRNA co-immunoprecipitated with FMRP and FXR2P and co-localized with these proteins in axons in the olfactory bulb (Table 2). As expected from the olfactory sensory neuron-specific expression of OMP, we did not detect Omp mRNA expression outside the olfactory system (data not shown). Mammalian axons thus contain ribonucleoprotein particles (RNPs) – the FXGs – that associate with specific FMRP target transcripts. Figure 3. Open in new tabDownload slide FXGs associate with mRNA in the juvenile mouse. (A) Schematic indicating location of depicted FXGs. Confocal micrographs of P15 mouse brains show that FXGs (identified by FXR2P) associate with (B) polyA+ RNA; (C-D) β-catenin mRNA, which colocalizes with rRNA; and (E) OMP mRNA. Quantification of this colocalization is in Table 2. For all panels, arrows: colocalizing FXGs; arrowheads: non-colocalizing FXGs; double arrowheads: non-FXG granules. Scale bar = 5 μm. Figure 3. Open in new tabDownload slide FXGs associate with mRNA in the juvenile mouse. (A) Schematic indicating location of depicted FXGs. Confocal micrographs of P15 mouse brains show that FXGs (identified by FXR2P) associate with (B) polyA+ RNA; (C-D) β-catenin mRNA, which colocalizes with rRNA; and (E) OMP mRNA. Quantification of this colocalization is in Table 2. For all panels, arrows: colocalizing FXGs; arrowheads: non-colocalizing FXGs; double arrowheads: non-FXG granules. Scale bar = 5 μm. In addition to those candidates tested above, biochemical and genetic studies have identified a large set of FMRP targets in the brain. We therefore asked whether there is a selective association of these transcripts with FMRP in axons (Table 1). Fmr1, Fxr1 and Fxr2 mRNAs were readily detected in neuronal cell bodies, but were not associated with FXGs (Supplementary Material, Fig. S3C–E). Other FMRP targets encoding presynaptic proteins such as piccolo, bassoon and the neurexins did not localize to FXGs (Table 1). We also examined the RNA encoding CaMKII-α, which is locally synthesized at nascent presynaptic sites (26), but did not detect it in FXGs (Table 1). As a further control for the sensitivity and specificity of the localization methods used here we assessed the distribution of a known postsynaptic FMRP target, Dlg4 (which encodes PSD-95). Dlg4 was readily detected in dendrites, but did not localize to FXGs in the same neuropil (Supplementary Material, Fig. S3J). Taken together these observations indicate that only a select subpopulation of FMRP targets localizes to the axonal FXGs. Species-dependent expression of axonal RNA granules in the adult rodent brain We next asked whether FXGs and translational machinery are expressed in adult brain axons. We first assessed forebrain FXG expression in mice and rats at multiple stages including juveniles (P16), sexually mature animals (P70) and adults (P150). The overall regional distribution of FXGs was similar between juvenile rats and mice (data not shown), although the abundance within individual brain regions differed between these species (Fig. 4A). Consistent with previous results (15), FXGs were largely restricted to the olfactory bulb in the adult mouse forebrain (Fig. 4A–D, Supplementary Material, Fig. S4A–C,G–I). However, to our surprise FXGs persisted in multiple adult rat forebrain regions, most notably the hippocampus (Fig 4A, E–G, Supplementary Material, Fig. S4D–F, J–L). These adult rat hippocampal FXGs are present in both the dentate mossy fibres and CA3 associational fibres. Further, these adult FXGs associate with ribosomes, polyA+ RNA, and β-catenin mRNA (Fig. 4H–J, Supplementary Material, Fig. S4M–O). This difference in adult expression between species did not reflect differences in genetic background: we observed similar results across seven mouse strains and two rat strains (C57Bl/6, CFW, CD1, FVB, 129, BALB/C, DBA2; and Sprague Dawley and Long-Evans, respectively; data not shown). Taken together, these findings show that adult mammalian axons contain mRNAs and translational machinery and that this expression is species-dependent. Figure 4. Open in new tabDownload slide FXGs exhibit species-dependent expression in the adult brain. (A) FXG abundance in selected rat and mouse brain regions across postnatal development; n = 5 for each age/species. FXG density in frontal cortex declined with age in both species (2-way ANOVA; species P = 0.9952, age P < 0.0001, interaction P = 0.0013). In mouse frontal cortex, FXG densities were: P16, 0.124 ± 0.017 FXGs/100 μm2; P70, 0.012 ± 0.003; P150, 0.015 ± 0.008. For rats, the FXGs densities were: P16, 0.077 ± 0.006; P70, 0.037 ± 0.005; P150, 0.038 ± 0.011. Olfactory sensory neuron FXG expression varied between species but remained high in adults of both species (2-way ANOVA; species P = 0.0052, age P = 0.0168, interaction P = 0.0109). For mouse olfactory bulb, FXG densities were: P16, 1.963 ± 0.146; P70, 1.484 ± 0.143; P150, 1.181 ± 0.218. For rat olfactory bulb, FXG densities were: P16, 1.799 ± 0.261; P70, 2.832 ± 0.210; P150, 1.720 ± 0.335. Hippocampal mossy fibre FXG density declined with age in mice, but remained high in rats (2-way ANOVA; species P < 0.0001, age P = 0.0693, interaction P < 0.0001). For mouse hippocampus, FXG densities were: P16, 0.385 ± 0.038; P70, 0.039 ± 0.008; P150, 0.041 ± 0.009. In rats, hippocampal FXG densities were: P16, 0.273 ± 0.029; P70, 0.562 ± 0.072; P150, 0.434 ± 0.035. (B-G) FXGs decline with age in (B-D) the mouse hippocampus, but remain abundant across the lifespan in (E-G) the rat hippocampus as visualized in mossy fibres (MF). (H-J) FXGs in adult rat hippocampal mossy fibres (MF; H,I) and CA3 associational fibres (AF; J) contain ribosomes (H,J) and RNA (I,J) including the Ctnnb1 mRNA (J). For all panels, arrows: FXGs. Scale bar = 15 μm in B-G, 10 μm in H-J. Figure 4. Open in new tabDownload slide FXGs exhibit species-dependent expression in the adult brain. (A) FXG abundance in selected rat and mouse brain regions across postnatal development; n = 5 for each age/species. FXG density in frontal cortex declined with age in both species (2-way ANOVA; species P = 0.9952, age P < 0.0001, interaction P = 0.0013). In mouse frontal cortex, FXG densities were: P16, 0.124 ± 0.017 FXGs/100 μm2; P70, 0.012 ± 0.003; P150, 0.015 ± 0.008. For rats, the FXGs densities were: P16, 0.077 ± 0.006; P70, 0.037 ± 0.005; P150, 0.038 ± 0.011. Olfactory sensory neuron FXG expression varied between species but remained high in adults of both species (2-way ANOVA; species P = 0.0052, age P = 0.0168, interaction P = 0.0109). For mouse olfactory bulb, FXG densities were: P16, 1.963 ± 0.146; P70, 1.484 ± 0.143; P150, 1.181 ± 0.218. For rat olfactory bulb, FXG densities were: P16, 1.799 ± 0.261; P70, 2.832 ± 0.210; P150, 1.720 ± 0.335. Hippocampal mossy fibre FXG density declined with age in mice, but remained high in rats (2-way ANOVA; species P < 0.0001, age P = 0.0693, interaction P < 0.0001). For mouse hippocampus, FXG densities were: P16, 0.385 ± 0.038; P70, 0.039 ± 0.008; P150, 0.041 ± 0.009. In rats, hippocampal FXG densities were: P16, 0.273 ± 0.029; P70, 0.562 ± 0.072; P150, 0.434 ± 0.035. (B-G) FXGs decline with age in (B-D) the mouse hippocampus, but remain abundant across the lifespan in (E-G) the rat hippocampus as visualized in mossy fibres (MF). (H-J) FXGs in adult rat hippocampal mossy fibres (MF; H,I) and CA3 associational fibres (AF; J) contain ribosomes (H,J) and RNA (I,J) including the Ctnnb1 mRNA (J). For all panels, arrows: FXGs. Scale bar = 15 μm in B-G, 10 μm in H-J. RNA granules are expressed in circuit-integrated axons The persistent expression of the translational machinery in axons of adult rat CA3 neurons, which are born exclusively in the prenatal period (27), indicates that axonal translation functions in mature, circuit-integrated axons. However, this observation does not resolve whether FXG-associated translational machinery is also present in growing axons prior to synaptogenesis. To address this question we examined olfactory sensory neurons, where growing and circuit-integrated axons can be distinguished from each other by the expression of GAP43 and OMP, respectively (28,29). As shown in Fig. 5B and C and Supplementary Material, Fig. S5B, FXGs are detected in axons expressing OMP, but are not observed in GAP43-positive axons. Moreover, FXGs are not observed in degenerating axons (Supplementary Material, Fig. S5C). These results are consistent with observations in hippocampus, where FXGs are not detected until P15 (15) – a time well after the initial formation of the axonal projections. These results demonstrate that FXGs are restricted to mature, circuit-integrated axons and are not a feature of growing axons. Figure 5. Open in new tabDownload slide Ribosomes and mRNA are expressed in axons of mature, circuit-integrated neurons. (A) Schematic indicating location of depicted FXGs. (B-C) FXGs were not detected in immature (GAP43-expressing) olfactory sensory neuron axons in adult mice in olfactory nerve layer tracts (ONL; B) or in glomerular neuropil (GL; C). (D-G) Ablation of neurogenesis in GFAP-TK transgenic adult rats by valganciclovir administration for 10 weeks had no effect on FXG density (control: 0.401 ± 0.033 FXGs/100 μm2; neurogenesis-deficient: 0.444 ± 0.040; unpaired t-test: P = 0.474; n = 5 rats for each condition). For all panels, arrows: colocalizing FXGs; arrowheads: non-colocalizing FXGs. Scale bar = 5 μm in A-B; 10 μm in E-F. Figure 5. Open in new tabDownload slide Ribosomes and mRNA are expressed in axons of mature, circuit-integrated neurons. (A) Schematic indicating location of depicted FXGs. (B-C) FXGs were not detected in immature (GAP43-expressing) olfactory sensory neuron axons in adult mice in olfactory nerve layer tracts (ONL; B) or in glomerular neuropil (GL; C). (D-G) Ablation of neurogenesis in GFAP-TK transgenic adult rats by valganciclovir administration for 10 weeks had no effect on FXG density (control: 0.401 ± 0.033 FXGs/100 μm2; neurogenesis-deficient: 0.444 ± 0.040; unpaired t-test: P = 0.474; n = 5 rats for each condition). For all panels, arrows: colocalizing FXGs; arrowheads: non-colocalizing FXGs. Scale bar = 5 μm in A-B; 10 μm in E-F. The expression of axonal RNA granules in adult hippocampus is independent of ongoing neurogenesis In the dentate gyrus newly born granule cells are generated throughout life and exhibit a period of heightened plasticity as they integrate into the mature circuitry (30–32). Adult neurogenesis is more robust in rats than mice (33). Thus it is possible that the interspecies differences in FXG expression in the hippocampus (Figs. 4, Supplementary Material, Fig. S4) could reflect the selective expression of FXGs in axons of adult-born granule cells during this window of enhanced plasticity. We tested this possibility using three complementary methods to manipulate adult neurogenesis in rats. First, we ablated adult neurogenesis by administering valganciclovir to three-month old transgenic GFAP-TK rats that express herpes simplex thymidine kinase in hippocampal neural progenitors (34). As shown in Fig. 5D–G, FXG abundance in mossy fibres was unchanged following a 10-week blockade of adult neurogenesis (P = 0.2369, one-tailed t-test). Second, FXG abundance was insensitive to X-irradiation blockade of neurogenesis (Supplementary Material, Fig. S5D and E). Third, increasing neurogenesis via electroconvulsive seizures had no detectable effect on FXG expression up to 28 days after treatment (Supplementary Material, Fig. S5F–M). These results demonstrate that FXGs are expressed in mossy fibres of neurons that had been generated early in life. However, we note that since adult-born neurons contribute only ∼6% of total dentate neurons per month (35), these observations do not rule out the possibility that adult-generated mossy fibres may also express FXGs. Taken together with the observations in CA3 associational fibres described above, these data establish that translational machinery is expressed within adult axons of mature, circuit-integrated neurons that are born early in life. Axonal transport of mRNA and ribosomes is independent of FMRP We next sought to elucidate the function of axonal FMRP. As an RNA binding protein that also regulates protein synthesis, FMRP could play a role in the axonal localization and/or the translational regulation of its target mRNAs. To test for a potential function in RNA localization we examined brains from P15–P23 Fmr1 null mice. As shown in Fig. 6B–F and Table 2, ribosomes and RNA, including transcripts encoding the FMRP targets β-catenin and OMP, localize to axons and associate with FXGs in these mutant brains. FMRP is thus not required for the axonal trafficking of translational machinery. However, there was a decrease in FXG association with ribosomes in the Fmr1 null FXGs to approximately two-thirds the association seen in the wild type FXGs (Table 2). Consistent with findings in the wild type mice, we observed that all FXGs (15/15) that colocalized with Y10b immunofluorescence in widefield micrographs exhibited direct association in super resolution micrographs (Fig. 6C). We also observed a decrease in the association of Omp mRNA with FXGs in Fmr1 null mice as compared to those in wild type mice (Table 2). However, as there are approximately four-fold more FXGs in axons of these mice (15), this likely represents an overall increase in the absolute number of ribosome- and Omp-associated FXGs in these mice. To more quantitatively examine axonal RNA transport, we took advantage of the distinctive organization of the olfactory system. Omp is expressed exclusively by the sensory neurons, whose cell bodies reside in the olfactory epithelium within the nasal cavity, while their axons project to the olfactory bulb in the brain. Thus Omp mRNA levels can be measured in both somatic (olfactory epithelium) and axonal (olfactory bulb) preparations from the same circuit. As shown in Fig. 6G, qPCR analysis showed that neither the overall expression nor the axonal targeting of Omp mRNA was sensitive to loss of FMRP. Finally, to completely ablate FXGs we examined mice that do not express FXR2P at P30 (15,36). As shown in Supplementary Material, Fig. S6B neither the overall expression nor axonal localization of Omp mRNA requires FXR2P or FXGs. Thus, both somatic expression and axonal trafficking of ribosomes and the target mRNAs are independent of FMRP and FXR2P. Figure 6. Open in new tabDownload slide FMRP regulates axonal protein expression of FXG targets independent of RNA expression or localization. (A) Schematic indicating location of images. In brains from Fmr1 null mice, FXGs associate with: (B-C) ribosomes; (D) polyA+ RNA; (E)Ctnnb1 mRNA; and (F)Omp mRNA. Quantification of the colocalization is in Table 2. (G)Omp mRNA expression and axonal localization in Fmr1 null mice are comparable to those in controls. For somatic expression in olfactory epithelium, qPCR values were 1.451 ± 0.621 in wild type and 1.610 ± 0.798 in Fmr1 null mice. For axonal localization in olfactory bulb, values were 1.292 ± 0.444 in wild type and 1.122 ± 0.431 in Fmr1 nulls. n = 5 for each genotype. (H)Fmr1 null mice exhibit increased axonal expression of OMP protein (wild type: 0.870 ± 0.109; Fmr1 null: 1.617 ± 0.238) but not GAP43 protein (wild type: 1.474 ± 0.403; Fmr1 null: 1.072 ± 0.199). (I-J) Representative images of olfactory glomeruli depicting OMP and GAP43 proteins filling olfactory sensory neuron axons as they innervate the olfactory bulb from wild type and Fmr1 null mice. n = 5 for each genotype. For all panels, arrows: colocalizing FXGs; arrowheads: non-colocalizing FXGs; double arrowheads: non-FXG granules. Scale bar = 5 μm for B,D-F; 450 nm for C; 35 μm for I-J. Figure 6. Open in new tabDownload slide FMRP regulates axonal protein expression of FXG targets independent of RNA expression or localization. (A) Schematic indicating location of images. In brains from Fmr1 null mice, FXGs associate with: (B-C) ribosomes; (D) polyA+ RNA; (E)Ctnnb1 mRNA; and (F)Omp mRNA. Quantification of the colocalization is in Table 2. (G)Omp mRNA expression and axonal localization in Fmr1 null mice are comparable to those in controls. For somatic expression in olfactory epithelium, qPCR values were 1.451 ± 0.621 in wild type and 1.610 ± 0.798 in Fmr1 null mice. For axonal localization in olfactory bulb, values were 1.292 ± 0.444 in wild type and 1.122 ± 0.431 in Fmr1 nulls. n = 5 for each genotype. (H)Fmr1 null mice exhibit increased axonal expression of OMP protein (wild type: 0.870 ± 0.109; Fmr1 null: 1.617 ± 0.238) but not GAP43 protein (wild type: 1.474 ± 0.403; Fmr1 null: 1.072 ± 0.199). (I-J) Representative images of olfactory glomeruli depicting OMP and GAP43 proteins filling olfactory sensory neuron axons as they innervate the olfactory bulb from wild type and Fmr1 null mice. n = 5 for each genotype. For all panels, arrows: colocalizing FXGs; arrowheads: non-colocalizing FXGs; double arrowheads: non-FXG granules. Scale bar = 5 μm for B,D-F; 450 nm for C; 35 μm for I-J. FMRP regulates axonal protein expression FMRP is well established as a negative regulator of mRNA translation in neuronal somata and dendrites, with loss of FMRP resulting in elevated expression of many of the proteins encoded by its target mRNAs (37). The association of axonal FMRP with mRNA and ribosomes suggested that it could play such a regulatory role in this neuronal domain as well. To test this prediction, we assessed OMP protein expression in somatic and axonal compartments of wild type and Fmr1 null mice. As shown in Fig. 6H–J and Supplementary Material, Fig. S6C and D, axonal OMP levels are increased in Fmr1 null compared to wild type mice. To quantify this increase, we performed within-animal normalization of axonal OMP protein expression levels to somatic expression levels and compared this ratio among animals. Since OMP expression can be influenced by an animal’s olfactory experience (38), this method controlled for any effect on OMP levels that might reflect differences in the odor environment among cages. This approach also isolated axonal effects by controlling for any global effect on OMP expression that might arise from the loss of FMRP. As shown in Fig. 6H there was over a 50% increase in the relative axonal protein expression in Fmr1 null as compared to control mice. In contrast, there was no detectable change in the relative expression of GAP43. Taken together, these results indicate that FMRP negatively regulates the axonal protein expression of its targets. Axonal RNA granules are expressed in the adult human brain We next investigated whether the translational machinery is present in adult human brain axons by asking whether FXGs are expressed in human hippocampus. We examined human hippocampal sections from a 44-year old male and a 57-year old female. We observed FXG expression in the hippocampus of each of these individuals (Fig. 7, Supplementary Material, Fig. S7). As in mice and rats, these FXGs contain both FXR2P and FMRP (Fig. 7A and B, E and F). The spatial pattern of FXG expression in rodent hippocampus was also conserved in human hippocampus: FXGs were expressed in mossy fibre (Figs. 7A–D, Supplementary Material, S7A) and CA3 associational fibre (Fig. 7E–H) axons but were not observed in Schaffer collaterals (Supplementary Material, Fig. S7B). The tissue of the 44-year old was particularly well preserved. We therefore investigated whether FXGs in this individual associated with translational machinery. As in the rat, these adult human FXGs contain both ribosomes (Fig. 7C and G) and polyA+ RNA (Fig. 7D and H). Notably, the FXGs in CA3 associational fibres are expressed in neurons that were born and synaptically integrated into circuits many decades earlier. Thus, translational machinery including ribosomes, RNA and translational regulators is a feature of axons in the adult human brain. Figure 7. Open in new tabDownload slide Axonal mRNA and ribosomes in FXGs of the adult human brain. Confocal micrographs of hippocampal sections from a 44-year-old individual. In these sections, FMRP and FXR2P localize to both FXGs (arrows) and neuronal cell bodies (arrowheads). (A-D) FXGs are found in adult human mossy fibre axons where they contain FXR2P as well as FMRP (A-B), ribosomes (C), and polyA+ RNA (D). (E-H) FXGs in CA3 associational fibres contain FXR2P as well as FMRP (E,F), ribosomes (G) and polyA+ RNA (H). Scale bar = 10 μm in A,E; 5 μm in B-D,F-H. Figure 7. Open in new tabDownload slide Axonal mRNA and ribosomes in FXGs of the adult human brain. Confocal micrographs of hippocampal sections from a 44-year-old individual. In these sections, FMRP and FXR2P localize to both FXGs (arrows) and neuronal cell bodies (arrowheads). (A-D) FXGs are found in adult human mossy fibre axons where they contain FXR2P as well as FMRP (A-B), ribosomes (C), and polyA+ RNA (D). (E-H) FXGs in CA3 associational fibres contain FXR2P as well as FMRP (E,F), ribosomes (G) and polyA+ RNA (H). Scale bar = 10 μm in A,E; 5 μm in B-D,F-H. Discussion In this report, we demonstrate that mature axons in the adult mammalian brain contain ribosomes, RNA and translational regulators including FMRP. Importantly, our findings establish the presence of endogenous RNA-associated granules – FXGs – in axons that are integrated into intact, native circuitry. Both FXG circuit-selectivity and the association with ribosomes and mRNA are conserved across the three mammalian species examined. FXG-target RNAs include transcripts encoding the plasticity-related proteins β-catenin and OMP. Our findings support the model that FMRP provides local regulatory control of axonal translation of associated FXG targets. The presence of FXGs in adult humans supports a role for FMRP-regulated axonal translation throughout life. Dysregulated axonal plasticity due to the absence of such control may contribute to symptoms seen in FXS patients. FXGs and axonal translation The approach used here takes advantage of the distinctive morphology and composition of the FXG to unambiguously identify axonal structures across the CNS. This approach, which was validated by specifically manipulating axonal FMRP expression in defined brain regions (Fig. 1, Supplementary Material, Fig. S1), permits the use of a comprehensive marker set to detect ribosomes and mRNA in FXGs and assign them to mature CNS axons. Axonal ribosomes were detected using in situ hybridization probes for 18S and 28S rRNA as well as antibodies for 5S/5.8S rRNA and the S6 ribosomal protein. Associated mRNAs were characterized by in situ hybridization probes for polyA+ RNA and for specific transcripts. The specificity of the in situ hybridization was supported both by standard controls as well as the region-selective association of specific mRNAs with FXGs. For example, Omp mRNA was only observed in olfactory sensory neurons, whereas Ctnnb1 mRNA was detected in a subset of FXGs in several brain regions. Taken together, these findings establish that ribosomes and specific transcripts, associated with a discrete granule, are present in mature CNS axons. The presence of translational machinery in axons strongly suggests that axons synthesize proteins in the brain. In support of this role, we observed that axonal protein levels of the FMRP target OMP are elevated in Fmr1 mutants. This dysregulation of axonal protein expression in the absence of FMRP is unlikely to reflect changes in mRNA transcription, transport or stability: Fmr1 null mice exhibit no detectable change in levels or axonal localization of ribosomes or FXG targets such as Omp mRNA. Given the established role of FMRP as a negative regulator of translation (37) the elevated axonal OMP in Fmr1 null mice is most likely to result from increased protein synthesis. In principle, the increased axonal OMP could be locally synthesized or may be due to increased somatic protein synthesis followed by selective axonal transport of the additional OMP. However, since the axonal OMP is increased relative to that in the cell bodies (Fig. 6, Supplementary Material, Fig. S6), such a mechanism would necessitate that FMRP specifically regulate the post-translational localization of OMP. We are not aware of any evidence for such a role for FMRP. Therefore, we favour the simplest model where FMRP regulates axonal OMP expression at the level of local translational control within axons. This model is strongly supported by the colocalization within axonal granules of large and small ribosome subunits, Omp mRNA and translational regulators including the Fragile X proteins. Taken together with the results discussed above, these findings suggest that FMRP locally regulates axonal translation in the intact brain. Our findings reveal that a highly restricted repertoire of FMRP target mRNAs is present in axons. Only 4 of 21 validated FMRP target transcripts examined were associated with FXGs (Table 1). In contrast, all 21 mRNAs were readily detected in cell bodies. Moreover, Ctnnb1 was the sole mRNA studied that robustly associated with FXGs in multiple forebrain regions. These findings suggest that FMRP regulates the translation of distinct sets of transcripts in different subcellular domains within the same neuron. This subcellular specificity, combined with the restriction of FXGs to specific neuron types (14,15), points to an exquisitely precise regulation of axonal transport and translation of mRNAs in the mammalian brain. FXG structure Structured illumination super resolution microscopy showed that FXGs are roughly 200 nm in size, a similar dimension to Drosophila P bodies detected by similar imaging methods (39). However, it should be noted that the limits of structured illumination microscopy constrain this measure to an approximate upper bound. The super resolution analysis further revealed that FXG components are enriched in subdomains of the FXG, with overlapping but distinct distributions of FXR2P, FMRP, ribosomes and mRNA within the granule (Figs. 2 and 6, Supplementary Material, Figs S2 and S4). Distinct subdomains are observed in several classes of non-neuronal RNA granules, suggesting that common structural strategies underpin the organization of these diverse cytoplasmic assemblages (39–42). Dynamic association of RNA with FXGs Our results suggest that the association of FXGs and translational machinery is dynamic. Axonal Omp mRNA is associated with both FXGs and non-FXG complexes. Further, almost half of FXGs in all brain regions studied lack detectable RNA or ribosomes. Moreover, FMRP and FXR2P are observed in ribosome-free domains of FXGs (Figs. 2D, 6C), suggesting these proteins can self-assemble into FXGs. Consistent with this organization, FXR2P and FMRP both contain low complexity domains, which can mediate the formation of granule-like droplets in the absence of RNA (43). Taken together, these observations support a model in which FXGs form independently of RNA and then dynamically associate with translational machinery. One intriguing possibility is that FXGs are associated with ion channels, since FMRP directly binds to ion channels and modulates their function in axons (11,44–46). These findings suggest a model in which FXGs constitutively associate with channels while mRNAs transition between FXG and non-FXG complexes in response to signalling events. In this model, FXG-associated mRNAs would be translationally repressed by FMRP. These mRNAs would then be actively translated upon dissociating from FXGs. Such dynamic association of FXGs with mRNA and ribosomes could serve to couple membrane potential and translation within select axons. The dynamic association of RNA and FXGs also suggests that there are discrete mechanisms for the transport and the translational regulation of axonal transcripts. Axonal transport of ribosomes and RNA does not require FMRP, and the FMRP target Omp mRNA does not require FXGs for its localization to axons. The mechanisms that direct the axonal targeting and assembly of RNPs in the brain are largely unknown. However, the signals that guide these processes in axons must differ from those that operate in dendrites, since FXGs localize to restricted populations of axons while dendritic RNPs are widespread throughout the nervous system. Finally, FXR2P may be particularly important for axonal RNA regulation. FXR2P is the sole member of the Fragile X family that is N-terminally myristoylated, a modification that modulates the axonal distribution of FXR2P (47). Since FXR2P is also the only Fragile X protein required for FXG expression (15), FXR2P myristoylation may regulate FXG localization. Together, these findings suggest a model in which FXGs, ribosomes and RNA are separately transported into axons where their dynamic assembly into FXR2P-dependent RNPs is regulated by local cues. FXGs in the adult brain Local protein synthesis in axons has been best studied in developing and regenerating systems, particularly in the context of growth cone guidance (7). However, we found no evidence that FXGs localize to growing axons. Instead, FXGs are present only after axons become synaptically-integrated into circuits. Further, by manipulating adult neurogenesis, we determined that FXGs persist for months after the period of robust plasticity in newly born adult neurons (30–32). Therefore, we propose that FXGs regulate local translation in the mature axonal arbour, likely in the context of adult plasticity. In support of such a role, mammalian brain axons contain transcripts encoding the plasticity-related proteins β-catenin, a regulator of synaptic vesicle distribution (48,49) and OMP, which modulates odor-induced signal transduction (50–53). FXGs in the adult hippocampus exhibit remarkable species-dependence – persisting in rats and humans but disappearing before P60 in mice. Interestingly, rats inhabit larger and more varied habitats and exhibit a greater behavioural repertoire than mice. For example, mice are exclusively terrestrial while rats are also arboreal and aquatic (54,55). Rats also show superior performance across a broader array of spatial learning tasks, exhibit better spatial learning retention, and are more adept at using flexible spatial strategies to solve problems (56–60). One intriguing hypothesis is that axonal translation in the adult rat hippocampus supports the formation and experience-dependent modification of broad territory maps required to navigate larger and more varied habitats. This mechanism may also contribute to the immense behavioural plasticity of humans. Dysregulation of translation in mature axons could contribute to behavioural and neurological deficits in Fragile X syndrome and perhaps autism. FXGs are enriched in neurons in sensorimotor and cortical circuits that mediate functions that are especially perturbed in FXS patients (14). Since FXG-containing neurons lose FMRP in both the somatodendritic and axonal domains, these neurons are likely particularly susceptible to loss of this protein in FXS. Importantly, the axonal trafficking of ribosomes and FMRP target mRNAs persists in these neurons in Fragile X mice. Therefore, dysregulated local protein synthesis could result in altered axonal dosage of FMRP targets important for presynaptic plasticity (Fig. 8). Targeting the axonal FMRP translational control pathway in Fragile X patients could correct abnormal presynaptic function. Finally, the large majority of neurodevelopmental and neurodegenerative diseases affect selected brain circuits. However, most candidate therapeutic targets for these disorders are expressed in all neurons if not all cell types. The exquisite circuit selectivity of the FXG thus presents a spatially – and functionally – precise target for therapeutic intervention. Figure 8. Open in new tabDownload slide A model for FMRP-regulated axonal translation. Axons in the central nervous system contain mRNA and ribosomes that associate with FXGs. In FXG-containing neurons, local translation contributes to the axonal proteome. In the absence of FMRP, these neurons contain FXGs that associate with ribosomes and mRNA but exhibit dysregulated axonal translation. This dysregulation results in a cell type-specific increase in the dosage of select presynaptic proteins that mediate axonal plasticity. Figure 8. Open in new tabDownload slide A model for FMRP-regulated axonal translation. Axons in the central nervous system contain mRNA and ribosomes that associate with FXGs. In FXG-containing neurons, local translation contributes to the axonal proteome. In the absence of FMRP, these neurons contain FXGs that associate with ribosomes and mRNA but exhibit dysregulated axonal translation. This dysregulation results in a cell type-specific increase in the dosage of select presynaptic proteins that mediate axonal plasticity. In this study we investigated a structure expressed in a subset of neurons, the FXG, as a means to characterize translational machinery in mature adult axons. However, it seems likely that axonal translation is a general phenomenon that is regulated by different mechanisms in other neurons. Indeed, axonal translation might be induced in some neurons only in response to specific physiological or pathological cues (4,61). In support of this idea, retinal ganglion cell axons harbour ribosome-associated mRNAs in the adult nervous system yet do not contain FXGs (5,14). Identification of specific RNA granules in other axonal populations promises to inform our understanding of mechanisms regulating translation in mature axons in the normal and diseased mammalian brain. Materials and methods Animals All work with animals was performed in accordance with protocols approved by the Institutional Animal Care and Use Committee of Brown University, Drexel University, the Rockefeller University and/or the National Institutes of Health. Male and female animals did not exhibit differences in FXG expression or composition at any age. All studies used a mix of males and females, except for the neurogenesis (electroconvulsive seizure and GFAP-TK) experiments, which only used males. C57BL/6 wild-type mice, Fmr1 knockout mice, and Fxr2 knockout mice as well as Sprague Dawley rats were deeply anaesthetized by isoflurane inhalation before intracardiac perfusion with room temperature HBS (0.1M HEPES, pH 7.4; 150 mM sodium chloride) containing 1 U/ml heparin and 0.5% sodium nitrite followed by perfusion with room temperature PBS (0.1M phosphate, pH 7.4; 150 mM sodium chloride) containing 4% paraformaldehyde. After perfusion, intact brains were carefully removed and postfixed overnight in the perfusate. After washing in PBS, brains were transferred to PBS containing 30% sucrose until the brains sank. Brains were then embedded in OCT medium by rapid freezing and stored at −80 °C until sectioned. Free-floating coronal or sagittal sections of OCT-embedded brains were prepared using a Leica cryostat at 40 μm and either used the same day or stored at −20 °C in antifreeze solution (50 mM phosphate buffer, pH 7.4; 30% sucrose; 30% ethylene glycol; 1% polyvinyl pyrrolidone) until use for immunolabelling. For in situ hybridization, animals were euthanized by isoflurane inhalation. Brains were rapidly dissected, embedded in OCT, frozen at −80 °C, and stored until sectioning. Human tissue Human brain specimens were collected with full approval of the institutional review boards at the Yale School of Medicine and at each institution or brain bank collecting the tissue, and with appropriate written informed consent from next of kin. The handling of tissue was performed in accordance with ethical guidelines and regulations for the research use of human tissue set forth by the NIH and the WMA Declaration of Helsinki. These samples are associated with no codes, links, or identifiers that could be used to reidentify the source subjects. Tissue was fixed using PBS containing 4% paraformaldehyde, sectioned, and stained. Electroconvulsive seizure (ECS) rats Eight-week-old male Sprague-Dawley rats (Charles River Laboratories, Wilmington, MA) were pair housed and maintained in standard conditions with a 12-h light/dark cycle and ad libitum access to food and water. Rats were habituated to the colony room for at least one week before ECS or sham treatment was administered. ECS was delivered as described (62) via ear-clip electrodes by using a pulse generator (Ugo Basile, Varese, Italy; 60 mA, 0.5-sec duration, 100-Hz frequency) to induce a generalized grand mal seizure lasting for <15 s. Sham groups were handled identically but received no shock. Animals were immediately returned to their home-cages where they remained until they were sacrificed. Each time point and the sham group consisted of three animals. GFAP-TK rats. Rats expressing HSV-TK (thymidine kinase from herpes simplex virus) under the GFAP promoter (GFAP-TK rats) were generated on a Long-Evans background and genotyped by PCR as previously described for GFAP-TK mice (63). Male wild type and transgenic littermates were genotyped after weaning and pair housed under a standard 12-h light/dark cycle. Beginning at 12 weeks of age, five rats of each genotype were given valganciclovir for 10 weeks (100 mg/kg/week, per orem, for 2 weeks, then 20 mg/kg/week thereafter) to eliminate neurogenesis in GFAP-TK rats but not wild type controls (34). Rats were then transcardially perfused with 4% paraformaldehyde, and brains were sectioned at 40 μm on a sliding microtome before processing for immunostaining. Antibodies. For detection of the ribosomal RNA 5S and 5.8S subunits, we used supernatant from hybridoma cells expressing the monoclonal antibody Y10b (gift from Dr. J. Twiss; 1:1000) (64,65). For detection of FXR2P, we used the ascites form of the monoclonal 1G2 (Developmental Studies Hybridoma Bank; 1:1000), the monoclonal A42 (Sigma; 1:500) or the monoclonal 55 (BD Biosciences; 1:500). For detection of FMRP, we used the monoclonal 2F5-1 (5 μg/ml) (15) or a rabbit polyclonal (Abcam ab17722; 1:500). Additional antibodies used included those against OMP (Wako; 1:1000), GAP43 (Novus NB300-143; 1:1000), neurofilament (Millipore AB1981; 1:1500) and S6 (Cell Signaling; 1:500). Secondary antibodies (Life Technologies) conjugated to Alexa fluors were used at 1:1000. Immunofluorescence in rodents Tissue stored in antifreeze was first washed three times in PBS (10 mM phosphate pH 7.4; 150 mM NaCl) before antigen retrieval, while tissue stained immediately after cutting was directly transferred into antigen retrieval solution. In order to improve antibody accessibility to epitopes in tissue, sections were first heated in 0.01M sodium citrate (pH 6.0) for 30 min at 75 °C. Tissue was then treated with blocking solution [PBST (10 mM phosphate buffer, pH 7.4, and 0.3% Triton X-100) and 1% blocking reagent (Roche)] for 30 min to occupy nonspecific binding sites. Sections were then treated with blocking solution plus primary antibody overnight before being washed for 5 min with PBST. For secondary detection, tissue was incubated with appropriate labelled antibodies in blocking solution for 1 h. Tissue was then washed for 5 min with PBST, mounted in 4% n-propylgallate, 60% glycerol, 5 mM phosphate pH 7.4, then coverslipped and sealed with nail polish. Sections were imaged using a Zeiss LSM 510 confocal microscope. Super-resolution microscopy Super-resolution microscopy was performed either on samples generated as above (supplemental figures) or on CLARITY-treated (66) tissue sections (main figures). In this protocol, mice were perfused, brains were extracted, and sections were prepared as above. These sections were stored in PBS containing 0.02% sodium azide at 4 °C until further processing. For CLARITY, sections were incubated in a modified HydroGel containing PBS, 4% PFA, 4% acrylamide, and 0.25% VA-044 (Wako) at 4 °C overnight. These sections were then incubated at 37 °C for 3 h in the same hydrogel in order to crosslink the tissue to the matrix. Tissue was then incubated in clearing solution (200 mM borate, pH 8.5, and 4% sodium dodecyl sulphate) for approximately two weeks and then immunostained using antibodies diluted in 0.5M sodium borate and 0.1% Triton X-100. Tissue sections were incubated in primary antibodies overnight, washed twice for 10 min each in borate/triton, incubated in secondary antibodies for 2 h, washed twice more for 10 min each in borate/triton, and then incubated in FocusClear (Cedarlane) overnight. These sections were mounted on a cover glass in 4% n-propylgallate, 85% glycerol, and 5 mM phosphate pH 7.4. Super-resolution structured illumination microscopy imaging was performed using a Deltavision OMX in the Drexel University Department of Biology Cell Imaging Center. Quantification of OMP and GAP43 immunofluorescence Brains and the nasal cavity from perfusion-fixed animals were dissected prior to postfixing as above. After postfixing, tissue was decalcified in 0.5M EDTA, pH 8 for 4 days and then cryoprotected overnight in PBS + 30% sucrose. Tissue was then embedded in OCT and frozen and stored at −80°C until serial cryosectioning at 20 μm for slide mounting using a Leica cryostat. Tissue was stored at −20°C until processing for immunofluorescence. To quantify OMP and GAP43 immunofluorescence, five sections from the main olfactory bulb and five sections from the main olfactory epithelium were imaged for each animal using a Hamamatsu Orca ER camera coupled to a Nikon Eclipse T800 microscope using a 20X objective. In each image, the intensity of OMP or GAP43 signal across a defined area of a glomerulus in the olfactory bulb or olfactory sensory neuron cell bodies in the olfactory epithelium was measured. Background, the average pixel value in an unstained region of tissue within the same image, was subtracted from the measured axonal or somatic intensity. For each animal, the cell body and the axonal intensity of OMP and GAP43 was determined by averaging each of the five images. To assess the subcellular distribution of these proteins, a ratio of the axonal intensity of both OMP and GAP43 in the axonal compartment over the cell body compartment was determined for each animal. Data are presented as the mean ± standard error of this axon/soma ratio on a per animal basis. Images for presentation were collected on a Zeiss LSM 510 scanning confocal microscope at 63x. All images for quantification were collected and analysed by workers blind to genotype. RNase Treatment of Tissue Fresh frozen sections from mouse brain were allowed to warm to room temperature, before fixation for 10’ with PBS prewarmed to 37 °C containing 4% PFA. Sections were rinsed three times with PBS and then incubated for 3 h at 37 °C in either TE (10 mM Tris, pH 8.0 + 1 mM EDTA) or TE + RNase (Ambion RNase cocktail @ 1:10; Ambion RNase V1 @ 1:10). Tissue was then rinsed three times in PBS, and processed for immunohistochemistry for FXR2P and rRNA as above. Immunofluorescence on human tissue Tissue was washed 3X in PBS before treatment in 0.01M sodium citrate (pH 6.0) for 30’ at 75 °C. Tissue was then treated with blocking solution for 30’ before incubation overnight in blocking solution plus A42 anti-FXR2P and 2F5-1 anti-FMRP. The following day, sections were rinsed for 5’ in PBST, incubated in blocking solution plus appropriate secondaries, and rinsed once more in PBST for 5’. To decrease lipofuscin fluorescence (67), sections were treated in 1 mM copper sulphate in 50 mM ammonium acetate (pH 5.0) for 10’, rinsed 1X in distilled water, then rinsed 5X in PBS prior to mounting in NPG. Sections were imaged using a Zeiss LSM 510 confocal microscope. The CuSO4 treatment greatly reduced lipofuscin fluorescence; remaining lipofuscins were excluded based on their red fluorescence under ultraviolet (DAPI filter set) illumination. In situ hybridization for polyA+ mRNA This protocol was adapted from previously published methods (68,69). In brief, for juvenile mice fresh frozen brains were sectioned at 20 μm, mounted on slides, and stored at −20 °C until use. All subsequent steps were performed at room temperature unless otherwise noted. On the day of staining, slides were allowed to warm to room temperature, fixed in PBS + 4% PFA for 10 min, and washed 3X in PBS. Sections were treated in 0.01M sodium citrate, pH 6 for 30’ at 75 °C followed by incubation for 10 min in 0.2M HCl and 2’ in PBS + 1% Triton X-100. Sections were then rinsed twice for 1 min each in PBS before a 10-min equilibration in 2X SSC + 10% formamide. Sections were exposed overnight at 37 °C to hybridization solution (10% dextran sulphate, 1 mg/ml E. coli tRNA, 2 mM Vanadyl Ribonucleoside, 200 μg/ml BSA, 2X SSC, 10% deionized formamide) containing 80 nM oligo(dT)45 that had been end-labelled with DIG Oligonucleotide Tailing Kit, 2nd generation (Roche) following the included instructions for short tails. On the second day, the slides were washed two times for 30’, each at 37 °C in 2X SSC + 10% formamide, and then rinsed briefly in 2X SSC followed by a rinse in PBST. Primary antibodies against digoxigenin (sheep polyclonal from Roche at 1:200) and FXR2P (A42 from Sigma at 1:500) were applied in blocking solution for 2 h. Tissue was rinsed before secondary antibody application in blocking solution for 1 h. Sections were rinsed and then mounted in mounting medium. Sections were imaged using a Zeiss LSM 510 confocal. In situ hybridization for individual mRNAs The first day of the protocol was as for oligo(dT) probes above. For hybridization, sections were incubated overnight at 37 °C in hybridization solution containing 250 nM Stellaris fluorophore-labelled oligonucleotide probes (Biosearch Technologies; sequences in Supplementary Material, Table S2) along with the A42 monoclonal anti-FXR2P. Sections were then rinsed two times for 30’ each at 37 °C in 2X SSC + 10% formamide followed by 2X SSC and then PBST. Sections were then incubated in blocking solution plus a secondary antibody for 1 h to detect A42, washed three times in PBST, mounted in NPG, and imaged using a Zeiss LSM 510 confocal. Super-resolution structured illumination microscopy imaging was performed using a Deltavision OMX in the Drexel University Department of Biology Cell Imaging Center. In situ hybridization in adult tissue In situ for Ctnnb1 was performed on slide-mounted fresh frozen sections as above. In situ hybridization for polyA+ RNA was performed as above except using free-floating sections from adult rat and human brains that were fixed prior to the in situ protocol by perfusion (rat) or immersion (human). This modified protocol began with a rinse in PBS and then progressing from the sodium citrate treatment onward. RNA Immunoprecipitation Wild type mice were sacrificed and their olfactory epithelia obtained via dissection. Tissue was homogenized in polysome lysis buffer as described (22). Lysate was precleared with protein A beads (Dynabeads, Life Technologies) for 1 h at 4 °C. Immunoprecipitation was performed with the following antibodies pre-bound to protein A beads: rabbit anti-FMRP (Abcam 1772), rabbit anti-FXR2P (BU38) (14), mouse anti-rRNA (Y10b, specifically binds 5S and 5.8S rRNA), mouse or rabbit IgG. Samples were incubated with beads for 2 h at 4 °C. Half of each sample was run on 10% polyacrylamide gels and transferred to nitrocellulose. Blots were probed for FMRP (2F5-1; 1:400) or FXR2P (BD Biosciences 611330; 1:5,000). RNA was extracted from the other half of each sample using the Trizol reagent (Life Technologies) for RNA. RNA amounts were quantified with the Quant-iT RiboGreen RNA Assay kit (Life Technologies) and reverse transcribed to cDNA using the iScript Kit (BioRad). The presence of OMP mRNA in each sample was assessed by PCR using the primer pair GGGAGAAGAAGCAGGATGGTGAGA and ATACATGACCTTGCGGATCTTGGC (70). Quantitative RT-PCR Littermate wild type and Fmr1 knockout mice were sacrificed at P23 and the olfactory bulbs and epithelium were separately dissected out. Similarly, P30 wild type mice and Fxr2 knockout mice were sacrificed and the olfactory bulbs and epithelium were removed. Five animals were used for each genotype. RNA was extracted using the Trizol reagent (Life Technologies), DNase treated (Turbo DNase, Life Technologies) and cleaned with RNeasy mini clean up kit (Qiagen). 200ng of RNA were used to reverse transcribe cDNA. qPCR using primers against OMP and GAPDH (Qiagen; QT00257544 and QT01658692, respectively) was performed on an Applied Biosystems 7300 Real Time PCR System. Total relative levels of OMP were determined by ΔΔCt normalizing to GAPDH and wild type levels. Data are depicted as mean ± standard error. Quantification of FXG identification Mice harbouring either a Cre-sensitive Rosa26 tdTomato reporter allele (Jackson Laboratory; Bar Harbor, ME) (19) or a Cre-sensitive Fmr1 allele (20) were crossed to mice in which Cre expression is driven by the Synapsin 1 promoter (Jackson Laboratory) or to mice in which Cre expression is driven specifically in olfactory sensory neurons by the #123 promoter (16–18). Previous characterization of the Synapsin Cre mouse line indicates that, rather than the expected pan-neuronal expression, this transgene drives Cre expression in populations of neurons including dentate and CA3 pyramidal neurons in hippocampus, neurons in thalamic relay nuclei, and a subpopulation of layer IV neurons in neocortex (21,14). Confocal micrographs of FMRP and FXR2P immunostaining from wild type (Synapsin-Cre;Fmr1wt/y or #123-Cre;Fmr1wt/y) and conditional knockout (Synapsin-Cre;Fmr1flox/y or #123-Cre;Fmr1flox/y) mice collected without pixel saturation were processed using the FIJI build of ImageJ. FXGs were identified in the FXR2P channel. These images were first processed with a rolling ball background subtraction with a radius of 7 pixels followed by a median filter with a radius of 2 pixels. Images were manually thresholded to highlight FXGs by including 0.05% of the total pixels. FXGs defined by the algorithm served as regions of interest for measuring the signal in the FMRP channel. An intensity in the FMRP channel was measured per FXG, and the area of each FXG was also measured. Background measurements were also collected to establish the signal expected by chance. For these background analyses, the image masks representing the regions of interest were rotated 180° and FMRP intensity was measured in the same images used for measuring signal in FXGs. This approach allowed for measuring signal in a way that was uncorrelated with the original image and yet represented measurements of the same size distribution and spatial frequency as found for the actual FXGs. Somatic regions of interest were established by automatically thresholding the FXR2P channel using the built-in Moments algorithm. Somatic FMRP signal and region of interest area were measured for all regions greater then 300 pixels2. FMRP intensity in wild type and conditional knockout ROIs was analysed by ANCOVA in R, using FMRP background intensity for each ROI as a covariate and controlling for multiple observations per image and animal. Data are depicted as mean ± standard error. Quantification of FXG colocalization Confocal micrographs collected without pixel saturation were processed using the FIJI build of ImageJ. Channels corresponding to the various FXG components were processed individually. For highlighting of FXGs, the images were processed with a rolling ball background subtraction with a radius of 7 pixels followed by a median filter with a radius of 2 pixels. The images were then manually thresholded to highlight FXGs, which resulted in including roughly 0.05% of total pixels in the processed image. To determine whether FXGs associated with each ribosome or RNA component, we analysed fluorescence intensity as compared to background in the appropriate channel of the image using FXG ROIs as described in the preceding paragraph. The fluorescence signal was compared to background fluorescence by ANOVA using Graphpad Prism. A second analysis determined how many FXGs contained the relevant component. Individual FXGs were found to be roughly 5 pixels in area when processed in this manner. For association with ribosome components (rRNA or rpS6) or polyA+ RNA, the total number of FXGs was counted. These FXGs were considered to colocalize with ribosomes if they contained one or more pixels in the top 1% of all ribosome signals. For colocalization with specific mRNAs (Ctnnb1 or Omp), the RNA channels were processed as for the FXGs above to identify RNA puncta. FXGs were considered to colocalize with mRNA if they contained one or more pixels highlighted in the RNA analysis. Similarly, Omp mRNA granules were considered to colocalize with FXGs if they contained one or more pixels highlighted in the FXG analysis. For statistical analyses, it was assumed that the ∼5-pixel FXGs would contain a pixel in the top 1% of ribosome or polyA+ signal 5% of the time or that they would contain Ctnnb1 or Omp mRNA signal (each representing ∼0.05% of an image) 0.25% of the time. χ2 analyses were utilized to determine whether the observed colocalization deviated from the expected results using Graphpad Prism. Comparisons between observed colocalization values in tissue from wild type and Fmr1 null mice were performed using Fisher’s exact test in Graphpad Prism. Quantification of FXG abundance Epifluorescent micrographs of FXR2P-stained sections were processed as above to identify FXGs. The number of FXGs per image was counted, with 2–5 images (technical replicates) analysed for each brain region and animal. These were averaged to give an individual number that represented that animal/region. ANOVAs for each brain region across the age and among species were performed using Graphpad Prism. As appropriate, Bonferroni-corrected post hoc t-tests comparing each region at each time point between conditions (e.g., rat vs mouse) were performed in Graphpad Prism. Data are depicted as mean ± standard error. Supplementary Material Supplementary Material is available at HMG online. Acknowledgements We thank B. McKechnie, C. Schmiedel, and D. Le for technical assistance. Conflict of Interest statement. None declared. Funding This work was supported by National Institutes of Health grants R00 MH090237 to MRA; R00 MH096939 to KYK; P50 MH106934, U01 MH103339, R01 MH109904 to NS; and R01 HD052083 to JRF. HAC was funded by the Intramural Program of the NIH, National Institute of Mental Health. Grant Number: ZIAMH002784. References 1 Pelkey K.A. , McBain C.J. ( 2008 ) Target-cell-dependent plasticity within the mossy fibre-CA3 circuit reveals compartmentalized regulation of presynaptic function at divergent release sites . J. Physiol ., 586 , 1495 – 1502 . Google Scholar Crossref Search ADS PubMed WorldCat 2 Hörnberg H. , Holt C. ( 2013 ) RNA-binding proteins and translational regulation in axons and growth cones . Neurogenomics , 7 , 81. Google Scholar OpenURL Placeholder Text WorldCat 3 Korsak L.I.T. , Mitchell M.E., Shepard K.A., Akins M.R. ( 2016 ) Regulation of Neuronal Gene Expression by Local Axonal Translation . Curr. Genet. Med. Rep ., 4 , 16 – 25 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Baleriola J. , Walker C.A., Jean Y.Y., Crary J.F., Troy C.M., Nagy P.L., Hengst U. ( 2014 ) Axonally Synthesized ATF4 Transmits a Neurodegenerative Signal across Brain Regions . Cell , 158 , 1159 – 1172 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Shigeoka T. , Jung H., Jung J., Turner-Bridger B., Ohk J., Lin J.Q., Amieux P.S., Holt C.E. ( 2016 ) Dynamic Axonal Translation in Developing and Mature Visual Circuits . Cell , 10.1016/j.cell.2016.05.029. Google Scholar OpenURL Placeholder Text WorldCat 6 Akins M.R. , Berk-Rauch H.E., Fallon J.R. ( 2009 ) Presynaptic translation: stepping out of the postsynaptic shadow . Front. Neural Circuits , 3 , 17. Google Scholar Crossref Search ADS PubMed WorldCat 7 Jung H. , Yoon B.C., Holt C.E. ( 2012 ) Axonal mRNA localization and local protein synthesis in nervous system assembly, maintenance and repair . Nat. Rev. Neurosci ., 13 , 308 – 324 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Bassell G.J. , Warren S.T. ( 2008 ) Fragile X Syndrome: Loss of Local mRNA Regulation Alters Synaptic Development and Function . Neuron , 60 , 201 – 214 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Bhakar A.L. , Dölen G., Bear M.F. ( 2012 ) The Pathophysiology of Fragile X (and What It Teaches Us about Synapses) . Annu. Rev. Neurosci ., 35 , 417 – 443 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Darnell J.C. , Klann E. ( 2013 ) The translation of translational control by FMRP: therapeutic targets for FXS . Nat. Neurosci ., 16 , 1530 – 1536 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Myrick L.K. , Deng P.Y., Hashimoto H., Oh Y.M., Cho Y., Poidevin M.J., Suhl J.A., Visootsak J., Cavalli V., Jin P., et al. . ( 2015 ) Independent role for presynaptic FMRP revealed by an FMR1 missense mutation associated with intellectual disability and seizures . Proc. Natl. Acad. Sci ., 112 , 949 – 956 . Google Scholar Crossref Search ADS WorldCat 12 Suhl J.A. , Muddashetty R.S., Anderson B.R., Ifrim M.F., Visootsak J., Bassell G.J., Warren S.T. ( 2015 ) A 3′ untranslated region variant in FMR1 eliminates neuronal activity-dependent translation of FMRP by disrupting binding of the RNA-binding protein HuR . Proc. Natl. Acad. Sci ., 112 , E6553 – E6561 . Google Scholar Crossref Search ADS WorldCat 13 Collins S.C. , Bray S.M., Suhl J.A., Cutler D.J., Coffee B., Zwick M.E., Warren S.T. ( 2010 ) Identification of novel FMR1 variants by massively parallel sequencing in developmentally delayed males . Am. J. Med. Genet. A , 152A , 2512 – 2520 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Akins M.R. , LeBlanc H.F., Stackpole E.E., Chyung E., Fallon J.R. ( 2012 ) Systematic mapping of fragile X granules in the mouse brain reveals a potential role for presynaptic FMRP in sensorimotor functions . J. Comp. Neurol ., 520 , 3687 – 3706 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Christie S.B. , Akins M.R., Schwob J.E., Fallon J.R. ( 2009 ) The FXG: A Presynaptic Fragile X Granule Expressed in a Subset of Developing Brain Circuits . J Neurosci ., 29 , 1514 – 1524 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Hirata T. , Nakazawa M., Yoshihara S., Miyachi H., Kitamura K., Yoshihara Y., Hibi M. ( 2006 ) Zinc-finger gene Fez in the olfactory sensory neurons regulates development of the olfactory bulb non-cell-autonomously . Development , 133 , 1433 – 1443 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Takeuchi H. , Inokuchi K., Aoki M., Suto F., Tsuboi A., Matsuda I., Suzuki M., Aiba A., Serizawa S., Yoshihara Y., et al. . ( 2010 ) Sequential Arrival and Graded Secretion of Sema3F by Olfactory Neuron Axons Specify Map Topography at the Bulb . Cell , 141 , 1056 – 1067 . Google Scholar Crossref Search ADS PubMed WorldCat 18 Kaneko-Goto T. , Sato Y., Katada S., Kinameri E., Yoshihara S., Nishiyori A., Kimura M., Fujita H., Touhara K., Reed R.R., et al. . ( 2013 ) Goofy Coordinates the Acuity of Olfactory Signaling . J. Neurosci ., 33 , 12987 – 12996 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Madisen L. , Zwingman T.A., Sunkin S.M., Oh S.W., Zariwala H.A., Gu H., Ng L.L., Palmiter R.D., Hawrylycz M.J., Jones A.R., et al. . ( 2010 ) A robust and high-throughput Cre reporting and characterization system for the whole mouse brain . Nat. Neurosci ., 13 , 133 – 140 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Mientjes E.J. , Nieuwenhuizen I., Kirkpatrick L., Zu T., Hoogeveen-Westerveld M., Severijnen L., Rifé M., Willemsen R., Nelson D.L., Oostra B.A. ( 2006 ) The generation of a conditional Fmr1 knock out mouse model to study Fmrp function in vivo . Neurobiol. Dis ., 21 , 549 – 555 . Google Scholar Crossref Search ADS PubMed WorldCat 21 He X.P. , Kotloski R., Nef S., Luikart B.W., Parada L.F., McNamara J.O. ( 2004 ) Conditional Deletion of TrkB but Not BDNF Prevents Epileptogenesis in the Kindling Model . Neuron , 43 , 31 – 42 . Google Scholar Crossref Search ADS PubMed WorldCat 22 Darnell J.C. , Van Driesche S.J., Zhang C., Hung K.Y.S., Mele A., Fraser C.E., Stone E.F., Chen C., Fak J.J., Chi S.W., et al. . ( 2011 ) FMRP Stalls Ribosomal Translocation on mRNAs Linked to Synaptic Function and Autism . Cell , 146 , 247 – 261 . Google Scholar Crossref Search ADS PubMed WorldCat 23 Taylor A.M. , Berchtold N.C., Perreau V.M., Tu C.H., Li Jeon N., Cotman C.W. ( 2009 ) Axonal mRNA in Uninjured and Regenerating Cortical Mammalian Axons . J Neurosci ., 29 , 4697 – 4707 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Ressler K.J. , Sullivan S.L., Buck L.B. ( 1994 ) Information coding in the olfactory system: Evidence for a stereotyped and highly organized epitope map in the olfactory bulb . Cell , 79 , 1245 – 1255 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Vassar R. , Chao S.K., Sitcheran R., Nuñez J.M., Vosshall L.B., Axel R. ( 1994 ) Topographic organization of sensory projections to the olfactory bulb . Cell , 79 , 981 – 991 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Sebeo J. , Hsiao K., Bozdagi O., Dumitriu D., Ge Y., Zhou Q., Benson D.L. ( 2009 ) Requirement for Protein Synthesis at Developing Synapses . J. Neurosci ., 29 , 9778 – 9793 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Angevine J.B. ( 1965 ) Time of neuron origin in the hippocampal region. An autoradiographic study in the mouse . Exp. Neurol. Suppl ., 1 – 70 . Google Scholar OpenURL Placeholder Text WorldCat 28 Monti-Graziadei G.A. , Margolis F.L., Harding J.W., Graziadei P.P. ( 1977 ) Immunocytochemistry of the olfactory marker protein . J. Histochem. Cytochem. Off. J. Histochem. Soc ., 25 , 1311 – 1316 . Google Scholar Crossref Search ADS WorldCat 29 Verhaagen J. , Oestreicher A.B., Gispen W.H., Margolis F.L. ( 1989 ) The expression of the growth associated protein B50/GAP43 in the olfactory system of neonatal and adult rats . J. Neurosci ., 9 , 683 – 691 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 30 Gu Y. , Arruda-Carvalho M., Wang J., Janoschka S.R., Josselyn S.A., Frankland P.W., Ge S. ( 2012 ) Optical controlling reveals time-dependent roles for adult-born dentate granule cells . Nat. Neurosci ., 15 , 1700 – 1706 . Google Scholar Crossref Search ADS PubMed WorldCat 31 Marín-Burgin A. , Mongiat L.A., Pardi M.B., Schinder A.F. ( 2012 ) Unique Processing During a Period of High Excitation/Inhibition Balance in Adult-Born Neurons . Science , 335 , 1238 – 1242 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Nakashiba T. , Cushman J.D., Pelkey K.A., Renaudineau S., Buhl D.L., McHugh T.J., Barrera V.R., Chittajallu R., Iwamoto K.S., McBain C.J., et al. . ( 2012 ) Young Dentate Granule Cells Mediate Pattern Separation whereas Old Granule Cells Contribute to Pattern Completion . Cell , 149 , 188 – 201 . Google Scholar Crossref Search ADS PubMed WorldCat 33 Snyder J.S. , Choe J.S., Clifford M.A., Jeurling S.I., Hurley P., Brown A., Kamhi J.F., Cameron H.A. ( 2009 ) Adult-Born Hippocampal Neurons Are More Numerous, Faster Maturing, and More Involved in Behavior in Rats than in Mice . J. Neurosci ., 29 , 14484 – 14495 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Snyder J.S. , Grigereit L., Russo A., Seib D.R., Brewer M., Pickel J., Cameron H.A. ( 2016 ) A Transgenic Rat for Specifically Inhibiting Adult Neurogenesis . eNeuro , 3 , Google Scholar OpenURL Placeholder Text WorldCat 35 Cameron H.A. , McKay R.D. ( 2001 ) Adult neurogenesis produces a large pool of new granule cells in the dentate gyrus . J. Comp. Neurol ., 435 , 406 – 417 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Bontekoe C.J.M. , McIlwain K.L., Nieuwenhuizen I.M., Yuva-Paylor L.A., Nellis A., Willemsen R., Fang Z., Kirkpatrick L., Bakker C.E., McAninch R., et al. . ( 2002 ) Knockout mouse model for Fxr2: a model for mental retardation . Hum. Mol. Genet ., 11 , 487 – 498 . Google Scholar Crossref Search ADS PubMed WorldCat 37 Darnell J.C. , Richter J.D. ( 2012 ) Cytoplasmic RNA-Binding Proteins and the Control of Complex Brain Function . Cold Spring Harb. Perspect. Biol ., 4 , a012344. Google Scholar Crossref Search ADS PubMed WorldCat 38 Waguespack A.M. , Reems M.R., Butman M.L., Cherry J.A., Coppola D.M. ( 2005 ) Naris occlusion alters olfactory marker protein immunoreactivity in olfactory epithelium . Brain Res ., 1044 , 1 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Weil T.T. , Parton R.M., Herpers B., Soetaert J., Veenendaal T., Xanthakis D., Dobbie I.M., Halstead J.M., Hayashi R., Rabouille C., et al. . ( 2012 ) Drosophila patterning is established by differential association of mRNAs with P bodies . Nat. Cell Biol ., 14 , 1305 – 1313 . Google Scholar Crossref Search ADS PubMed WorldCat 40 West J.A. , Mito M., Kurosaka S., Takumi T., Tanegashima C., Chujo T., Yanaka K., Kingston R.E., Hirose T., Bond C., et al. . ( 2016 ) Structural, super-resolution microscopy analysis of paraspeckle nuclear body organization . J. Cell Biol ., 214 , 817 – 830 . Google Scholar Crossref Search ADS PubMed WorldCat 41 Wheeler J.R. , Matheny T., Jain S., Abrisch R., Parker R. ( 2016 ) Distinct stages in stress granule assembly and disassembly . eLife , 5 , e18413. Google Scholar OpenURL Placeholder Text WorldCat 42 Trcek T. , Grosch M., York A., Shroff H., Lionnet T., Lehmann R. ( 2015 ) Drosophila germ granules are structured and contain homotypic mRNA clusters . Nat. Commun ., 6 , 7962. Google Scholar Crossref Search ADS PubMed WorldCat 43 Conicella A.E. , Zerze G.H., Mittal J., Fawzi N.L. ( 2016 ) ALS mutations disrupt phase separation mediated by α-helical structure in the TDP-43 low-complexity C-terminal domain . Structure , 24 , 1537 – 1549 . Google Scholar Crossref Search ADS PubMed WorldCat 44 Brown M.R. , Kronengold J., Gazula V.R., Chen Y., Strumbos J.G., Sigworth F.J., Navaratnam D., Kaczmarek L.K. ( 2010 ) Fragile X mental retardation protein controls gating of the sodium-activated potassium channel Slack . Nat. Neurosci ., 13 , 819 – 821 . Google Scholar Crossref Search ADS PubMed WorldCat 45 Deng P.Y. , Rotman Z., Blundon J.A., Cho Y., Cui J., Cavalli V., Zakharenko S.S., Klyachko V.A. ( 2013 ) FMRP regulates neurotransmitter release and synaptic information transmission by modulating action potential duration via BK channels . Neuron , 77 , 696 – 711 . Google Scholar Crossref Search ADS PubMed WorldCat 46 Ferron L. , Nieto-Rostro M., Cassidy J.S., Dolphin A.C. ( 2014 ) Fragile X mental retardation protein controls synaptic vesicle exocytosis by modulating N-type calcium channel density . Nat. Commun ., 5 , 3628. Google Scholar Crossref Search ADS PubMed WorldCat 47 Stackpole E.E. , Akins M.R., Fallon J.R. ( 2014 ) N-myristoylation regulates the axonal distribution of the Fragile X-related protein FXR2P . Mol. Cell. Neurosci ., 62 , 42 – 50 . Google Scholar Crossref Search ADS PubMed WorldCat 48 Bamji S.X. , Shimazu K., Kimes N., Huelsken J., Birchmeier W., Lu B., Reichardt L.F. ( 2003 ) Role of β-catenin in synaptic vesicle localization and presynaptic assembly . Neuron , 40 , 719 – 731 . Google Scholar Crossref Search ADS PubMed WorldCat 49 Bamji S.X. , Rico B., Kimes N., Reichardt L.F. ( 2006 ) BDNF mobilizes synaptic vesicles and enhances synapse formation by disrupting cadherin–β-catenin interactions . J. Cell Biol ., 174 , 289 – 299 . Google Scholar Crossref Search ADS PubMed WorldCat 50 Reisert J. , Yau K.W., Margolis F.L. ( 2007 ) Olfactory marker protein modulates the cAMP kinetics of the odour-induced response in cilia of mouse olfactory receptor neurons . J. Physiol ., 585 , 731 – 740 . Google Scholar Crossref Search ADS PubMed WorldCat 51 Lee A.C. , He J., Ma M. ( 2011 ) Olfactory marker protein is critical for functional maturation of olfactory sensory neurons and development of mother preference . J. Neurosci ., 31 , 2974 – 2982 . Google Scholar Crossref Search ADS PubMed WorldCat 52 Kass M.D. , Moberly A.H., Rosenthal M.C., Guang S.A., McGann J.P. ( 2013 ) Odor-Specific, Olfactory Marker Protein-Mediated Sparsening of Primary Olfactory Input to the Brain after Odor Exposure . J. Neurosci ., 33 , 6594 – 6602 . Google Scholar Crossref Search ADS PubMed WorldCat 53 Dibattista M. , Reisert J. ( 2016 ) The odorant receptor-dependent role of olfactory marker protein in olfactory receptor neurons . J. Neurosci ., 36 , 2995 – 3006 . Google Scholar Crossref Search ADS PubMed WorldCat 54 Barnett S.A. ( 1975 ) The Rat;: A Study in Behavior University of Chicago Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 55 Berry R.J. , Bronson F.H. ( 1992 ) Life history and bioeconomy of the house mouse . Biol. Rev. Camb. Philos. Soc ., 67 , 519 – 550 . Google Scholar Crossref Search ADS PubMed WorldCat 56 Whishaw I.Q. ( 1995 ) A comparison of rats and mice in a swimming pool place task and matching to place task: Some surprising differences . Physiol. Behav ., 58 , 687 – 693 . Google Scholar Crossref Search ADS PubMed WorldCat 57 Whishaw I.Q. , Tomie J.A. ( 1996 ) Of mice and mazes: similarities between mice and rats on dry land but not water mazes . Physiol. Behav ., 60 , 1191 – 1197 . Google Scholar Crossref Search ADS PubMed WorldCat 58 Frick K.M. , Stillner E.T., Berger-Sweeney J. ( 2000 ) Mice are not little rats: species differences in a one-day water maze task . Neuroreport , 11 , 3461 – 3465 . Google Scholar Crossref Search ADS PubMed WorldCat 59 Stranahan A.M. ( 2011 ) Similarities and differences in spatial learning and object recognition between young male C57Bl/6J mice and Sprague-Dawley rats . Behav. Neurosci ., 125 , 791 – 795 . Google Scholar Crossref Search ADS PubMed WorldCat 60 Whishaw I.Q. , Metz G.A.S., Kolb B., Pellis S.M., Whishaw I.Q., Metz G.A.S., Kolb B., Pellis S.M. ( 2001 ) Accelerated nervous system development contributes to behavioral efficiency in the laboratory mouse: A behavioral review and theoretical proposal . Dev. Psychobiol. Dev. Psychobiol ., 39, 39 , 151, 151 – 170, 170 . Google Scholar OpenURL Placeholder Text WorldCat 61 Kalinski A.L. , Sachdeva R., Gomes C., Lee S.J., Shah Z., Houle J.D., Twiss J.L. ( 2015 ) mRNAs and protein synthetic machinery localize into regenerating spinal cord axons when they are provided a substrate that supports growth . J. Neurosci ., 35 , 10357 – 10370 . Google Scholar Crossref Search ADS PubMed WorldCat 62 Warner-Schmidt J.L. , Duman R.S. ( 2007 ) VEGF is an essential mediator of the neurogenic and behavioral actions of antidepressants . Proc. Natl. Acad. Sci ., 104 , 4647 – 4652 . Google Scholar Crossref Search ADS WorldCat 63 Snyder J.S. , Soumier A., Brewer M., Pickel J., Cameron H.A. ( 2011 ) Adult hippocampal neurogenesis buffers stress responses and depressive behaviour . Nature , 476 , 458 – 461 . Google Scholar Crossref Search ADS PubMed WorldCat 64 Lerner E.A. , Lerner M.R., Janeway C.A., Steitz J.A. ( 1981 ) Monoclonal antibodies to nucleic acid-containing cellular constituents: probes for molecular biology and autoimmune disease . Proc. Natl. Acad. Sci. U. S. A , 78 , 2737 – 2741 . Google Scholar Crossref Search ADS PubMed WorldCat 65 Garden G.A. , Canady K.S., Lurie D.I., Bothwell M., Rubel E.W. ( 1994 ) A biphasic change in ribosomal conformation during transneuronal degeneration is altered by inhibition of mitochondrial, but not cytoplasmic protein synthesis . J. Neurosci ., 14 , 1994 – 2008 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 66 Chung K. , Wallace J., Kim S.Y., Kalyanasundaram S., Andalman A.S., Davidson T.J., Mirzabekov J.J., Zalocusky K.A., Mattis J., Denisin A.K., et al. . ( 2013 ) Structural and molecular interrogation of intact biological systems . Nature , 497 , 332 – 337 . Google Scholar Crossref Search ADS PubMed WorldCat 67 Schnell S.A. , Staines W.A., Wessendorf M.W. ( 1999 ) Reduction of lipofuscin-like autofluorescence in fluorescently labeled tissue . J. Histochem. Cytochem ., 47 , 719 – 730 . Google Scholar Crossref Search ADS PubMed WorldCat 68 Akins M.R. , Greer C.A. ( 2006 ) Axon behavior in the olfactory nerve reflects the involvement of catenin-cadherin mediated adhesion . J. Comp. Neurol ., 499 , 979 – 989 . Google Scholar Crossref Search ADS PubMed WorldCat 69 Raj A. , van den Bogaard P., Rifkin S.A., van Oudenaarden A., Tyagi S. ( 2008 ) Imaging individual mRNA molecules using multiple singly labeled probes . Nat. Methods , 5 , 877 – 879 . Google Scholar Crossref Search ADS PubMed WorldCat 70 Oberto M. , Soncin I., Bovolin P., Voyron S., De Bortoli M., Dati C., Fasolo A., Perroteau I. ( 2001 ) ErbB-4 and neuregulin expression in the adult mouse olfactory bulb after peripheral denervation . Eur. J. Neurosci ., 14 , 513 – 521 . Google Scholar Crossref Search ADS PubMed WorldCat Published by Oxford University Press 2016. This work is written by US Government employees and is in the public domain in the US. Published by Oxford University Press 2016. This work is written by US Government employees and is in the public domain in the US.
TI - Axonal ribosomes and mRNAs associate with fragile X granules in adult rodent and human brains
JF - Human Molecular Genetics
DO - 10.1093/hmg/ddw381
DA - 2017-01-01
UR - https://www.deepdyve.com/lp/oxford-university-press/axonal-ribosomes-and-mrnas-associate-with-fragile-x-granules-in-adult-gUn0Vm2O0c
SP - 192
EP - 209
VL - 26
IS - 1
DP - DeepDyve
ER -