brain research 1596 (2015) 22–30
Available online at www.sciencedirect.com
www.elsevier.com/locate/brainres
Research Report
Expression of fragile X mental retardation protein in neurons and glia of the developing and adult mouse brain Shervin Gholizadeha,c, Sebok Kumar Haldera,c, David R Hampsona,b,n a
Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, and Department of Pharmacology, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada c S.H and S.G. contributed equally to this work b
ar t ic l e in f o
abs tra ct
Article history:
Fragile X syndrome is the most common inherited form of mental retardation and autism. It is
Accepted 10 November 2014
caused by a reduction or elimination of the expression of fragile X mental retardation protein
Available online 20 November 2014
(FMRP). Because fragile X syndrome is a neurodevelopmental disorder, it is important to fully
Keywords:
document the cell type expression in the developing CNS to provide a better understanding of the
Postnatal development
molecular function of FMRP, and the pathogenesis of the syndrome. We investigated FMRP
NG2
expression in the brain using double-labeling immunocytochemistry and cell type markers for
Neurons
neurons (NeuN), astrocytes (S100β), microglia (Iba-1), and oligodendrocyte precursor cells (NG2).
Astrocytes
The hippocampus, striatum, cingulate cortex, retrosplenial cortex, corpus callosum and cerebel-
Microglia
lum were assessed in wild-type C57/BL6 mice at postnatal days 0, 10, 20, and adult. Our results
Oligodendrocytes
demonstrate that FMRP is ubiquitously expressed in neurons at all times and brain regions
Oligodendrocyte precursor cells
studied, except for corpus callosum where FMRP was predominantly present in astrocytes at all
S100β
ages. FMRP expression in Iba-1 and NG2-positive cells was detected at postnatal day 0 and 10 and gradually decreased to very low or undetectable levels in postnatal day 20 and adult mice. Our results reveal that in addition to continuous and extensive expression in neurons in the immature and mature brain, FMRP is also present in astrocytes, oligodendrocyte precursor cells, and microglia during the early and mid-postnatal developmental stages of brain maturation. Prominent expression of FMRP in glia during these crucial stages of brain development suggests an important contribution to normal brain function, and in its absence, to the fragile X phenotype. & 2014 Elsevier B.V. All rights reserved.
1.
Introduction
of the FMR1 gene which induces a dramatic reduction or elimination of the expression of the encoded protein, fragile X
Fragile X syndrome (FXS) is a genetic disorder and a leading
mental retardation protein (FMRP). FMRP is an mRNA binding
cause of cognitive impairment and autism. The disorder is
protein that controls the expression of hundreds of genes in the
caused by triplet repeat expansion in the 5’ untranslated region
CNS through multiple mechanisms including modulating
n Correspondence to: Department of Pharmaceutical Sciences, University of Toronto, 144 College St. Toronto, Ontario, Canada M5S 3M2. E-mail address: [email protected] (D. Hampson).
http://dx.doi.org/10.1016/j.brainres.2014.11.023 0006-8993/& 2014 Elsevier B.V. All rights reserved.
23
brain research 1596 (2015) 22–30
ribosome stalling (Darnell et al., 2011). However, FMRP has been shown to also act as a positive modular of protein translation (Bechara et al., 2009), likely by enhancing mRNA stability (Zalfa et al., 2007). The FMR1 gene undergoes alternative splicing and possesses alternative transcription start sites such that at least 12 isoforms are generated (Brackett et al., 2013; Tassone et al., 2011). Isoform 1 is the longest form and codes for a protein of 632 amino acids and a molecular weight of 71kilodaltons. In the adult brain FMRP is abundantly expressed in most neurons throughout the CNS. In neurons, the protein is primarily located in the cytosol and in synaptic spines where it plays a role in spine maturation (Cruz-Martín et al., 2010). A portion of total cellular FMRP is also present in the nucleus where its role is less well characterized, and nuclear expression is isoform dependent (Dury et al., 2013). FMRP is highly abundant in “fragile X granules” in neuronal axons and presynaptic terminals where it apparently regulates recurrent neuronal activity (Akins et al., 2012). FMRP is also present in neural stem cells (Luo et al., 2010) where it has been shown to control hippocampal-dependent learning in the mature brain (Guo et al., 2011). In contrast to the expression of FMRP in neurons of the adult brain, relatively little is known about the types of glial cells that express FMRP during CNS development. FMRP was reported in primary cultures of rodent astrocytes (Yuskaitis et al., 2010), and in the mouse hippocampus, FMRP is expressed in astrocytes within the first week of birth and then declines to low or undetectable levels (Pacey and Doering, 2007). It is also present in oligodendroglia (Wang et al., 2004) and oligodendrocyte precursor cells (OPCs) in the immature cerebellum where it appears to be a factor in the proper progression of myelination (Pacey et al., 2013). The purpose of the present study was to more fully document the cell type expression of FMRP in neurons and glia in selected brain regions of the developing and mature mouse CNS. Our analysis encompassed the cerebral cortex, striatum, hippocampus, cerebellum, and the corpus callosum, and extended from postnatal days (PNDs) 0, to PNDs 10, 20, and adult. Our findings reveal that FMRP is extensively expressed in glia in the developing postnatal brain and that with the exception of the corpus callosum, glial expression declines at different rates in different brain regions to low levels in the adult brain.
2.
Results
Immunocytochemical analysis was used to map the expression of FMRP in selected brain regions at PND 0 (day of birth), 10, 20, and young adult (8–12 weeks old) wild-type C57/BL6 mice. Sections from cingulate cortex, hippocampus, striatum, corpus callosum, and cerebellum were immunolabeled for FMRP as well as a cell-type specific marker for neurons, astrocytes, oligodendrocyte precursor cells, and microglia. The 5C2 monoclonal antibody used here has previously been shown to be specific for FMRP (LaFauci et al., 2013). Brain sections treated with secondary antibody only showed no fluorescence. We quantified the total number of cells expressing FMRP per visual field, as well as cells co-expressing FMRP with one of the four cell-specific markers, and calculated the percent of total FMRPpositive cells that co-expressed each of the four cell markers in the cingulate cortex and corpus callosum (Table 1).
2.1. Gradual reduction in the number of FMRP-positive cells from the early postnatal period to adulthood FMRPþ cell counts in the cingulate cortex and corpus callosum revealed that the number of FMRPþ cells in both brain regions was highest at PND 0 (see “Total FMRPþ” column in Table 1). In the cingulate cortex there was a gradual decline over time such that the adult level was approximately 57% of that at birth (P0). In the corpus callosum there was a modest decline whereby adult mice showed about 85% of the level at P0. Overall, these observations are consistent with previous reports suggestion developmentally declining levels of FMRP protein, peaking at first postnatal week and declining thereafter (Lu et al., 2004; Pacey et al., 2013).
2.2. Predominant neuronal expression of FMRP throughout development At all postnatal times examined, FMRP expression was predominantly neuronal in all of the brain regions analyzed, except for corpus callosum (Fig. 1, Tables 1 and 2) where only a few NeuNþ cells were present and most FMRP expression was observed in S100βþ cells (Fig. 2, Table 2). Quantitative analysis of cell type-specific expression of FMRP revealed
Table 1 – Quantitative analysis of cell type-specific expression of FMRP in the cingulate cortex and corpus callosum. The analysis was performed by counting the FMRP-positive neurons (NeuN), astrocytes (S100β), microglia (Iba-1) and oligodendrocyte precursor cells (NG2) in the cingulate cortex and corpus callosum. The results are reported as the average number of total FMRP-positive cells per visual field, and the percentage of FMRP-positive cells that co-localized with NeuN, S100β, Iba-1, and NG2. Data are presented as mean7S.E.M. Standard error values smaller than 1.0 are not listed. Brain region
Age
Total FMRPþ
% NeuNþ
% S100β
% Iba-1
% NG2
Cingulate cortex
PND 0 PND 10 PND 20 Adult
15475 12375 8675 8877
85 8971 9471 9471
5 2 2 3
4 5 2 2
6 3 4 571
Corpus callosum
PND 0 PND 10 PND 20 Adult
8273 7573 7673 7073
2 2 2 2
27 3373 3372 2273
20 1973 1071 3
18 14 11 7
24
brain research 1596 (2015) 22–30
Fig. 1 – Expression of FMRP in neurons of the wild-type mouse brain. Coronal sections from postnatal days 0, 10, 20, and adult mice were double immunolabeled using anti-FMRP and anti-NeuN, and counterstained with DAPI (A–T). Co-localization of FMRP in NeuN-positive neurons (FMRP, green; NeuN, red; DAPI, blue) is shown in the cingulate cortex, hippocampus, striatum, corpus callosum, and cerebellum. Arrows indicate examples of co-localization. The heavy dashed lines delineate the boundaries of the corpus callosum. Scale bar¼20 μm.
85%–94% co-expression with NeuN in FMRPþ cells in the cingulate cortex. In stark contrast only 2% NeuN/FMRP coexpression was detected in the corpus callosum at all of the time points studied.
2.3. Developmental expression of FMRP in astrocytes, microglia and oligodendrocyte precursor cells We examined FMRP expression in astrocytes by double labeling with anti-FMRP and anti-S100β. FMRP and S100β were co-localized at PND 0 in cells in the cingulate cortex, striatum, hippocampus, corpus callosum, and cerebellum (Fig. 2, Tables 1 and 2). By PND 10, PND 20, and adult mouse brain, the population of FMRP/S100β -positive cells had declined drastically in cingulate cortex and hippocampus, (less than 3%). Compared to the cingulate cortex and hippocampus, the decline in FMRP/S100 β co-expression occurred more slowly in the striatum where substantial co-expression was seen at PND 10, PND 20, but not in adult mice. Persistent coexpression of FMRP/S100β was also observed in the cerebellum at PND0, 10, 20 and adult mice, although the number of FMRP/S100β cell gradually decreased in adult mouse brain. Interestingly, FMRP was co-expressed in many S100βþ cells in the corpus callosum. Of the brain regions analyzed, the corpus callosum showed by far the largest percentage of cells co-expressing of FMRP and S100β (Fig. 2, Table 2); more than 75% of S100β positive cells in the corpus callosum co-expressed
FMRP at all time points. This is an intriguing finding as this brain region is mainly composed of white matter. Approximately 30% of FMRP-expressing cells in the corpus callosum also expressed S100β at all developmental stages analyzed (Tables 1 and 2). The percent of S100β/FMRP-positive cells was maintained at a moderate level from PND 0 to adult mice, suggesting a continual expression of FMRP in astrocytes from early postnatal stages through adulthood in this brain region. In addition to the expression of S100β in astrocytes, S100β is also expressed by a population of oligodendrocyte precursor cells and oligodendrocytes (Hachem et al., 2005; Steiner et al., 2008). Therefore, two population of S100β immunopositive cells may exist, one representing astrocytes and one dedicated to oligodendrocyte precursor cells. To better characterize these two populations of cells, we also used NG2 staining as a specific marker for oligodendrocyte precursor cells. The NG2 proteoglycan is expressed in oligodendrocyte precursor cells, but not in mature oligodendrocytes, both in vitro and in developing rodent brain (Polito and Reynolds, 2005). FMRP was expressed in many NG2þ oligodendrocyte precursor cells in the cingulate cortex, striatum, hippocampus, corpus callosum, and cerebellum at PND 0 (Fig. 3, Tables 1 and 2). Thereafter, from PND 10 to adult mouse brain, FMRP expression in NG2-positive cells progressively declined in all of the brain regions examined, reaching low co-localization in adult mice in the cingulate cortex,
25
brain research 1596 (2015) 22–30
Table 2 – Semi-quantitative analysis of cell type-specific FMRP expression in the developing and adult mouse brain. The level of co-localization of FMRP with NeuN-positive neurons, S100β-positive astrocytes, Iba-1-positive microglia, and NG2positive oligodendrocyte precursor cells in mouse brain was assessed in the brain regions indicated at PND 0, PND 10, PND 20, and adults. The extent of co-localization is reported as: -, none;þlow; þþ moderate; and þþþ high. Brain region
Age of mice
FMRPþ/NeuNþ
FMRPþ/S100βþ
FMRPþ/Iba-1þ
FMRPþ/NG2þ
Cingulate cortex
PND 0 PND 10 PND 20 Adult
þþþ þþþ þþþ þþþ
þ þ þ þ
þ þ þ þ
þ þ þ þ
Corpus callosum
PND 0 PND 10 PND 20 Adult
þ þ þ þ
þþþ þþþ þþþ þþþ
þþþ þþþ þþ þ
þþþ þþ þþ þ
Striatum
PND 0 PND 10 PND 20 Adult
þþþ þþþ þþþ þþþ
þþ þ þ þ
þþ þ þ -
þþ þ þ þ
Hippocampus
PND 0 PND 10 PND 20 Adult
þþþ þþþ þþþ þþþ
þþ þ -
þþ þ -
þ þ þ -
Cerebellum
PND 0 PND 10 PND 20 Adult
þþþ þþþ þþþ þþþ
þ þ þ þ
þ þ -
þ þþ þ þ
Fig. 2 – Expression of FMRP in the astrocytes of mouse brain. Coronal sections from postnatal days 0, 10, 20, and adult mice were labeled using anti-FMRP, anti-S100β, and DAPI. FMRP in S100β -positive astrocytes (FMRP, green; S100β, red; DAPI, blue) are shown in the cingulate cortex, hippocampus, striatum, corpus callosum, and cerebellum. Arrows indicate examples of colocalization of FMRP in S100β-positive astrocytes, whereas dashed-squares indicate the absence of FMRP in astrocytes. The heavy dashed lines delineate the boundaries of the corpus callosum. Scale bar¼ 20 μm.
26
brain research 1596 (2015) 22–30
Fig. 3 – Expression of FMRP in oligodendrocyte precursor cells in the mouse brain. Coronal sections were labeled using antiFMRP, anti-NG2, and DAPI. Co-localization of endogenous FMRP in NG2-positive oligodendrocyte precursor cells (FMRP, green; NG2, red; DAPI, blue) is shown in the cingulate cortex, hippocampus, striatum, corpus callosum, and cerebellum. Arrows indicate the co-localization of FMRP in NG2-positive oligodendrocyte precursor cells, whereas dashed-squares indicate the absence of FMRP in oligodendrocyte precursor cells. The heavy dashed lines delineate the boundaries of the corpus callosum. Scale bar¼ 20 μm.
Fig. 4 – Expression of FMRP in microglia. Coronal sections were labeled with anti-FMRP and anti-Iba-1. Co-localization of FMRP in Iba-1-positive microglia (FMRP, green; Iba1, red; DAPI, blue) is shown. Arrows indicate examples of co-localization of FMRP in Iba-1-positive microglia, whereas dashed-squares indicate the absence of FMRP in microglia. Scale bar¼20 μm.
brain research 1596 (2015) 22–30
corpus callosum, striatum, and cerebellum and undetectable co-localization in the hippocampus (Table 2). FMRP expression in microglia was examined by assessing co-expression with Iba-1. FMRP/ Iba-1 double labeled cells were present in all brain regions analyzed at PND 0 (Fig. 4). At PND 0, more than 75% of the total NG2þ cells and Iba-1þ cells co-expressed FMRP in the all the brain regions analyzed. At PND 10 and PND 20, the number of FMRP/Iba-1 co-localized cells remained relatively constant in the cingulate cortex, corpus callosum, and striatum, but decreased in the hippocampus and cerebellum by PND 20. In the brain regions studied here, there was a gradual decrease in FMRP/Iba-1 co-expression after PND 20 with no detectable co-localization in adult mice.
3.
Discussion
Our results demonstrate a predominantly neuronal expression of FMRP in all of the brain regions analyzed, except the corpus callosum. Our data also reveal the presence of FMRP in cells expressing markers for astrocytes (S100β), microglia (Iba1) and oligodendrocyte precursor cells (NG2) in the developing brain. In general, glial expression of FMRP diminished to low or undetectable levels in the adult brain following different trajectories in different brain regions. FMRP expression was abundant in astrocytes at PND 0 in all of the brain regions analyzed; thereafter, expression in astrocytes showed a gradual decline in striatum and hippocampus, reaching negligible numbers of FMRPþ/S100βþ cells in the adult mouse brain. Pacey and Doering (2007) reported FMRP expression in cells expressing the astrocytic marker GFAP in the hippocampus and the ependymal cells surrounding the third ventricle at PND 1 and PND 7, but no FMRP/GFAP co-localization in the brains of adult (2 month-old) mice. Our results confirmed the findings of Pacey and Doering, (2007) in the hippocampus using a different astrocytic marker (S100β), and extended these findings to other brain regions including the cingulate cortex, corpus callosum, striatum, and cerebellum. Quantitative analysis of cell-type expression of FMRP indicated that more than 20% of FMRPþ cells in the corpus callosum co-expressed S100β and this co-localization rate remained relatively constant from PND 0 to adult mice. In contrast, there were few NeuNþ cells detected in the corpus callosum. Expression of FMRP in astrocytes could be of crucial importance in regulating the established functions of astrocytes in synapse formation and neural plasticity. More specifically, during the early postnatal weeks in rodents when astrocyte support of neuron growth and synapse formation is vital, the lack of FMRP could contribute to the abnormal dendritic morphology and synapse development seen in FXS (Jacobs and Doering, 2010). Astrocytes dynamically regulate synaptic transmission, partly by releasing soluble synaptogenic molecules such as the astrocyte-secreted matricellular proteins, including Hevin, and thrombospondin-1 (Jones and Bouvier, 2014, or neurotrophic factors such as Neurotrophin-3 (Yang et al., 2012), which are highly expressed within first few weeks of postnatal development and decrease as the brain matures. Interestingly, Hevin, also known as SPARC-like 1, and Neurotrophin-3 have been identified as an mRNA
27
substrates for FMRP (Darnell et al., 2011) and excessive neurotrophin-3, secreted from Fmr1 knockout astrocytes, has been suggested to contribute to abnormal neuronal dendritic development in Fmr1 knockout mouse model (Yang et al., 2012). Furthermore, thrombospondin-1 protein is reduced in the astrocytes of Fmr1 knockout mice and synaptic deficits are corrected by its protein replacement in Fmr1 knockout mouse hippocampal neurons (Cheng et al., 2014). These findings highlight the important role for FMRP expression in astrocytes during early postnatal weeks, which coincide with the peak of synaptogenesis. FMRP expression has been reported in oligodendrocyte precursors in vitro (Wang et al., 2004) and in the early postnatal cerebellum (Pacey et al., 2013). Our data extend these findings by reporting a more comprehensive analysis of FMRP/NG2 co-localization in different brain regions at multiple time points during postnatal development of the CNS. FMRP expression is also present in mature MBPþ oligodendrocytes of rodents as well as human oligodendrocytes (Giampetruzzi et al., 2013). FMRP was previously shown to bind to MBP mRNA and down-regulate its translation in vitro (Wang et al., 2004). Moreover, the lack of FMRP is associated with delayed myelination in the Fmr1 knockout mouse cerebellum (Pacey et al., 2013). In the present study, we observed a gradual decline in the number of FMRPþ cells co-expressing NG2 from PND 0 to adult mice; this pattern was observed in all of the brain regions examined, but was most prominent in the corpus callosum and striatum. A decline in FMRP levels during oligodendrocyte differentiation was previously reported in primary cultures of oligodendrocytes derived from neonatal rat brain (Wang et al., 2004). This gradual decline indicates that FMRP likely plays an important role in myelin formation during CNS maturation. FMRP has been previously reported in cultures of microglial cells (Yuskaitis et al., 2009; Yuskaitis et al., 2010). However, the findings reported here provide the first direct evidence for FMRP expression in microglia in brain tissue. Post mortem studies have identified neuroinflammation and glial activation in the brains of individuals with idiopathic autism (Fatemi et al., 2011; Tetreault et al., 2012), and other studies suggest a similar pathology may be present in fragile X syndrome (Jacobs et al., 2012). FMRP expression was reported in the BV2 microglial cell line (Yuskaitis and Jope, 2009). However, despite microglial expression of FMRP, there was no difference in the production of pro-inflammatory cytokines IL-6 and TNFα after acute lipopolysaccharide induced activation of microglia in wild-type or Fmr1 knockout mice. In the current study, FMRP expression in microglia was detected most prominently during the first 10 postnatal days and declined thereafter. Interestingly, the number of microglial cells in mice is immensely increased during the first two postnatal weeks when the vast majority (495%) of microglia are born (Alliot et al., 1999). These findings suggest a potential regulatory role for FMRP in the cell lineage and maturation of microglia during early development. However, the exact molecular function of FMRP in these cells and its contribution to behavioral abnormalities in Fmr1 knockout mice remains to be elucidated. Quantitative analysis of FMRPþ cells in the cingulate cortex and corpus callosum demonstrated the highest number of FMRPþ cells at PND 0 which diminished steadily throughout
28
brain research 1596 (2015) 22–30
brain development. The abundant FMRP expression at birth and during the first 1–2 postnatal weeks indicates that the functional requirement of FMRP in glia is highest during this critical period of early brain development. In the drosophila nervous system FMRP appears to be present sequentially, first in neuroblasts and then in glia, and glial expression was required for neuroblast reactivation (Callan et al., 2012). Similar to the mouse model, the expression of the drosophila FMRP homolog is developmentally regulated (Tessier et al., 2008) and its reintroduction during brain development rescues the dendritic impairment in the drosophila model (Gatto et al., 2009), indicating a causal relationship between FMRP expression during a distinct developmental time-window and normal synapse formation. In the mouse CNS, the most intense period of synaptogenesis (Zito and Svoboda, 2002) and the most severe dendritic spine abnormality in Fmr1 knockout brain were found to coincide at the end of the first postnatal week when abundant FMRP expression occurs (Nimchinsky et al., 2001). It has been shown that FMRP re-introduction specifically in CNS neurons using viral vector-mediated gene therapy was able to correct selected behavioral abnormalities in the Fmr1 knockout mouse (Gholizadeh et al., 2014). Conceivably, the reintroduction of FMRP in neurons could indirectly result in correction of disrupted neuron-to-astrocyte signaling in FXS at a critical time during early development when astrocyte support of neuronal growth and synapse formation is vital (Cheng, et al., 2012). Since glia play indispensable roles in synaptic development (Corty and Freeman, 2013), developmentally regulated FMRP expression in glia may influence synaptic development of adjacent neurons. Future efforts could be directed toward determining whether FMRP reintroduction in astrocytes alone, or concomitantly in both astrocytes and neurons, may provide enhanced benefit (correction of abnormal behaviors) over that seen when expressed in neurons only (Glolizadeh et al., 2014).
4.
Experimental procedures
4.1.
Animals
Wild-type C57BL/6 J mice were bred at the University of Toronto and used for all the experiments. Mice were kept in a room maintained at constant temperature (2172 1C) and relative humidity (5575%) with an automatic 12 h light/dark cycle with free access to standard laboratory diet and tap water. All animal experiments were carried out in accordance with the guidelines set out by the Canadian Council on Animal Care and were approved by the University of Toronto Animal Care Committee.
4.2.
Tissue collection and preparation
PND 10, PND 20, and adult (PND 55–65) wild-type mice were anaesthetized with a mixture of ketamine and xylazine (intraperitoneal, 80 and 8 mg/kg, respectively) and perfused transcardially with a solution of 0.1 M phosphate buffer saline (PBS, pH 7.4), followed by 4% paraformaldehyde (PFA, pH 7.4). The brains were removed and post-fixed overnight in 4% PFA at 4oC, and then transferred to 30% sucrose solution for cryoprotection. For PND 0, pups were decapitated, rinsed
with PBS and the skull was fixed in 4% PFA at room temperature for 2–4 h. The brain was then removed from the skull and further incubated in 4% PFA at 4oC for 20–24 h followed by 30% sucrose solution overnight at 4 1C. Once the brains sank in the sucrose, they were mounted in optimum cutting temperature solution (Sakura, Torrance, CA) and frozen at -80 1C until sectioning. For PND 10, PND 20 and adult mice, floating coronal brain sections were prepared at a thickness of 25 mm using a cryostat (Leica Microsystems, Wetzlar, Germany). For PND 0, both floating and thawmount sections were used (25 μm thick).
4.3.
Immunohistochemistry and confocal microscopy
To perform double fluorescence immunostaining, coronal brain sections were washed three times with Tris-buffered saline (TBS, pH 7.6) for 5 min each at room temperature (RT). The antigen retrieval assay was performed at all times in which sections were incubated with 0.8% sodium borohydride for 10 min at RT followed by washing once with TBS, and then incubated with 0.01 M citrate buffer (pH 6.0) in the water bath at 75 1C for 45 min. After cooling for 30 min, sections were washed three times with TBS and blocked for 1 h at RT with TBS containing and 0.1% Triton X-100, 5% goat serum (SigmaAldrich, St. Louis, MO, USA) and goat anti-mouse IgG (dilution 1:50; Sigma-Aldrich). Sections were washed four times in TBS for 5 min each with gentle rocking and then incubated overnight at 4 1C in the following primary antibodies diluted in TBS containing 5% goat serum: mouse monoclonal anti-FMRP 5c2 (1:1000) (LaFauci et al., 2013), rabbit polyclonal anti-NeuN (1:2000; Cat. # ab128886, Abcam, Cambridge, UK), rabbit polyclonal anti-S100β (1:1000; Cat. # ab868, Abcam, Cambridge, UK), rabbit polyclonal anti-Iba-1 (1:2000; Cat. # 019-19741, WAKO, Richmond, VA, USA), and rabbit polyclonal anti-NG2 (1:500; Cat. # AB5320, Millipore, Temecula, CA, USA). Sections were washed with TBS five times for 10 min each and then incubated with following secondary antibodies (diluted in TBS containing 5% goat serum) for 2 h at RT in the dark with gentle shaking: goat anti-mouse Alexa Fluor 488 and goat antirabbit Alexa Fluor 594 (diluted 1:1000; Jackson ImmunoResearch Laboratories, West Grove, PA, USA). The nuclei were visualized with DAPI (1:80000; Sigma-Aldrich). Sections were then washed in PBS (five times, 10 min each) with minimal exposure to light, mounted on glass slides, air dried, and then coverslipped with Prolong Gold antifade mounting medium (Invitrogen, Carlsbad, CA, USA). Representative tissue sections covering discrete brain regions including cingulate cortex, corpus callosum, striatum, hippocampus, and cerebellum were examined with a laser-scanning confocal microscope (Nikon A1; Nikon Instruments, Tokyo, Japan). The images were captured at 60X magnification and analyzed with Image J software (NIH, Bethesda, MD, USA) and NIS-Elements software (Nikon Instruments, Tokyo, Japan).
4.4.
Semi-quantitative and quantitative analysis
For comparative analysis of cell type-specific FMRP expression in the PND 0, PND 10, PND 20 and adult mouse brain, two scoring methods were used: a quantitative scoring system to count the FMRP-positive neurons (NeuN as marker), astrocytes (S100β as
brain research 1596 (2015) 22–30
marker), microglia (Iba-1 as marker) and oligodendrocyte precursor cells (NG2 as marker) in the cingulate cortex and corpus callosum, and a semi-quantitative scoring system to analyze the FMRP co-localization with NeuN, S100β, Iba-1, and NG2 in neurons, astrocytes, microglia, and oligodendrocyte precursor cells in the different brain regions including cingulate cortex, corpus callosum, striatum, hippocampus, and cerebellum. Briefly, photomicrographs made with the 60X objective lens were used to assess the number of FMRP-positive neurons, astrocytes, microglia, and oligodendrocyte precursor cells in the different brain regions. The number of cells in each bright field image was counted using specific cell markers with clearly visible nuclei by DAPI staining in random square fields (45000 mm 45000 μm) in the coronal brain sections (Table 2). Cell counting was performed using three consecutive coronal brain sections from each PND 0, PND 10, PND 20, and adult mouse (total of 9 sections from three mice at each age group) and reported as the average number of FMRP-positive cells per visual field, and the average percentage of FMRP-positive cells that co-localized with NeuN, S100β, Iba-1, and NG2 to obtain a mean value for each mouse. For semi-quantitative analysis, a scoring system was used to analyze the co-localization in FMRPþ cells in the cingulate cortex, hippocampus, striatum, corpus callosum, and cerebellum. Levels of co-localization are presented as the percentage of FMRPþ cells that co-localized with each of the cell-specific markers (shown in Table 2). The results were summarized as: -, no co-localization; þ, low co-localization; þþ, moderate co-localization; and þþþ, high co-localization.
Acknowledgments We thank Dr. Laura K.K. Pacey for helpful comments on the manuscript. This work was supported by an operating grant from the Canadian Institutes of Health Research and the Canadian Institutes of Health Research Training Program in Biological Therapeutics.
r e f e r e n c e s
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. Alliot, F., Godin, I., Pessac, B., 1999. Microglia derive from progenitors, originating from the yolk sac, and which proliferate in the brain. Brain Res. Dev. Brain Res. 117, 145–152. Bechara, E., Didiot, M., Melko, M., Davidovic, L., Bensaid, M., Martin, P., Castets, M., Pognonec, P., Khandjian, E., Moine, H., Bardoni, B., 2009. A novel function for fragile X mental retardation protein in translational activation. PLoS Biol 7, e16. Brackett, D.M., Qing, F., Amieux, P.S., Sellers, D.L., Horner, P.J., Morris, D.R., 2013. FMR1 transcript isoforms: association with polyribosomes; regional and developmental expression in mouse brain. PLoS One 8, e58296. Callan, M.A., Clements, N., Ahrendt, N., Zarnescu, D.C., 2012. Fragile X Protein is required for inhibition of insulin signaling and regulates glial-dependent neuroblast reactivation in the developing brain. Brain Res. 1462, 151–161. Cheng, C., Sourial, M., Doering, L.C., 2012. Astrocytes and developmental plasticity in fragile X. Neural Plast 2012, 197491.
29
Cheng, C., Zhu, L., Doering, L.C., 2014. Astrocyte-secreted TSP-1 corrects synaptic and spine phenotypes in the fragile X mouse model. Fourteenth International Fragile X Conference., Orange County, CA. Corty, M.M., Freeman, M.R., 2013. Cell biology in neuroscience: architects in neural circuit design: glia control neuron numbers and connectivity. J. Cell Biol. 203, 395–405. Cruz-Martı´n, A., Crespo, M., Portera-Cailliau, C., 2010. Delayed stabilization of dendritic spines in fragile X mice. J. Neurosci. 30, 7793–7803. Darnell, J., Van Driesche, S., Zhang, C., Hung, K., Mele, A., Fraser, C., Stone, E., Chen, C., Fak, J., Chi, S., Licatalosi, D., Richter, J., Darnell, R., 2011. FMRP stalls ribosomal translocation on mRNAs linked to synaptic function and autism. Cell 146, 247–261. Dury, A.Y., El Fatimy, R., Tremblay, S., Rose, T.M., Cote, J., De Koninck, P., Khandjian, E.W., 2013. Nuclear fragile X mental retardation protein is localized to Cajal bodies. PLoS Genet. 9, e1003890. Gholizadeh, S., Arsenault, J., Xuan, I.C., Pacey, L.K., Hampson, D.R., 2014. Reduced phenotypic severity following adenoassociated virus-mediated Fmr1 gene delivery in fragile X mice. Neuropsychopharmacology. Giampetruzzi, A., Carson, J.H., Barbarese, E., 2013. FMRP and myelin protein expression in oligodendrocytes. Mol. Cell Neurosci. 56, 333–341. Guo, W., Allan, A., Zong, R., Zhang, L., Johnson, E., Schaller, E., Murthy, A., Goggin, S., Eisch, A., Oostra, B., Nelson, D., Jin, P., Zhao, X., 2011. Ablation of Fmrp in adult neural stem cells disrupts hippocampus-dependent learning. Nat. Med 17, 559–565. Hachem, S., Aguirre, A., Vives, V., Marks, A., Gallo, V., Legraverend, C., 2005. Spatial and temporal expression of S100B in cells of oligodendrocyte lineage. Glia 51, 81–97. Jacobs, S., Doering, L.C., 2010. Astrocytes prevent abnormal neuronal development in the fragile x mouse. J. Neurosci. 30, 4508–4514. Jacobs, S., Cheng, C., Doering, L.C., 2012. Probing astrocyte function in fragile X syndrome. Results Probl. Cell Differ. 54, 15–31. Jones, E.V., Bouvier, D.S., 2014. Astrocyte-secreted matricellular proteins in CNS remodelling during development and disease. Neural Plast 2014, 321209. LaFauci, G., Adayev, T., Kascsak, R., Nolin, S., Mehta, P., Brown, W.T., Dobkin, C., 2013. Fragile X screening by quantification of FMRP in dried blood spots by a Luminex immunoassay. J. Mol. Diagn. 15, 508–517. Lu, R., Wang, H., Liang, Z., Ku, L., O’Donnell, W., Li, T., Warren, W., Feng, Y., S.T., 2004. The fragile X protein controls microtubuleassociated protein 1B translation and microtubule stability in brain neuron development. Proc. Natl. Acad. Sci. U. S. A 101, 15201–15206. Luo, Y., Shan, G., Guo, W., Smrt, R., Johnson, E., Li, X., Pfeiffer, R., Szulwach, K., Duan, R., Barkho, B., Li, W., Liu, C., Jin, P., Zhao, X., 2010. Fragile x mental retardation protein regulates proliferation and differentiation of adult neural stem/ progenitor cells. PLoS Genet 6, e1000898. Nimchinsky, E., Oberlander, A., Svoboda, K., 2001. Abnormal development of dendritic spines in FMR1 knock-out mice. J. Neurosci. 21, 5139–5146. Pacey, L.K.K., Doering, L.C., 2007. Developmental expression of FMRP in the astrocyte lineage: implications for fragile X syndrome. Glia 55, 1601–1609. Pacey, L.K.K., Xuan, I.C., Guan, S., Sussman, D., Henkelman, R.M., Chen, Y., Thomsen, C., Hampson, D.R., 2013. Delayed myelination in a mouse model of fragile X syndrome. Hum. Mol. Genet. 22, 3920–3930.
30
brain research 1596 (2015) 22–30
Polito, A., Reynolds, R., 2005. NG2-expressing cells as oligodendrocyte progenitors in the normal and demyelinated adult central nervous system. J. Anat. 207, 707–716. Steiner, J., Bernstein, H.G., Bogerts, B., Gos, T., Richter-Landsberg, C., Wunderlich, M.T., Keilhoff, G., 2008. S100B is expressed in, and released from, OLN-93 oligodendrocytes: Influence of serum and glucose deprivation. Neuroscience 154, 496–503. Tassone, F., De Rubeis, S., Carosi, C., La Fata, G., Serpa, G., Raske, C., Willemsen, R., Hagerman, P.J., Bagni, C., 2011. Differential usage of transcriptional start sites and polyadenylation sites in FMR1 premutation alleles. Nucleic Acids Res. 39, 6172–6185. Tetreault, N.A., Hakeem, A.Y., Jiang, S., Williams, B.A., Allman, E., Wold, B.J., Allman, J.M., 2012. Microglia in the cerebral cortex in autism. J. Autism. Dev. Disord. 42, 2569–2584. Wang, H., Ku, L., Osterhout, D., Li, W., Ahmadian, A., Liang, Z., Feng, Y., 2004. Developmentally-programmed FMRP expression in oligodendrocytes: a potential role of FMRP in regulating translation in oligodendroglia progenitors. Hum. Mol. Genet. 13, 79–89.
Yang, Q., Feng, B., Zhang, K., Guo, Y.Y., Liu, S.B., Wu, Y.M., Li, X.Q., Zhao, M.G., 2012. Excessive astrocyte-derived neurotrophin-3 contributes to the abnormal neuronal dendritic development in a mouse model of fragile X syndrome. PLoS Genet. 8, e1003172. Yuskaitis, C.J., Jope, R.S., 2009. Glycogen synthase kinase-3 regulates microglial migration, inflammation, and inflammation-induced neurotoxicity. Cell Signal. 21, 264–273. Yuskaitis, C.J., Beurel, E., Jope, R.S., 2010. Evidence of reactive astrocytes but not peripheral immune system activation in a mouse model of Fragile X syndrome. Biochim. Biophys. Acta. 1802, 1006–1012. Zalfa, F., Eleuteri, B., Dickson, K., Mercaldo, V., De Rubeis, S., di Penta, A., Tabolacci, E., Chiurazzi, P., Neri, G., Grant, S., Bagni, C., 2007. A new function for the fragile X mental retardation protein in regulation of PSD-95 mRNA stability. Nat. Neurosci. 10, 578–587. Zito, K., Svoboda, K., 2002. Activity-dependent synaptogenesis in the adult Mammalian cortex. Neuron 35, 1015–1017.
Available online at www.sciencedirect.com
www.elsevier.com/locate/brainres
Research Report
Expression of fragile X mental retardation protein in neurons and glia of the developing and adult mouse brain Shervin Gholizadeha,c, Sebok Kumar Haldera,c, David R Hampsona,b,n a
Department of Pharmaceutical Sciences, Leslie Dan Faculty of Pharmacy, and Department of Pharmacology, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada c S.H and S.G. contributed equally to this work b
ar t ic l e in f o
abs tra ct
Article history:
Fragile X syndrome is the most common inherited form of mental retardation and autism. It is
Accepted 10 November 2014
caused by a reduction or elimination of the expression of fragile X mental retardation protein
Available online 20 November 2014
(FMRP). Because fragile X syndrome is a neurodevelopmental disorder, it is important to fully
Keywords:
document the cell type expression in the developing CNS to provide a better understanding of the
Postnatal development
molecular function of FMRP, and the pathogenesis of the syndrome. We investigated FMRP
NG2
expression in the brain using double-labeling immunocytochemistry and cell type markers for
Neurons
neurons (NeuN), astrocytes (S100β), microglia (Iba-1), and oligodendrocyte precursor cells (NG2).
Astrocytes
The hippocampus, striatum, cingulate cortex, retrosplenial cortex, corpus callosum and cerebel-
Microglia
lum were assessed in wild-type C57/BL6 mice at postnatal days 0, 10, 20, and adult. Our results
Oligodendrocytes
demonstrate that FMRP is ubiquitously expressed in neurons at all times and brain regions
Oligodendrocyte precursor cells
studied, except for corpus callosum where FMRP was predominantly present in astrocytes at all
S100β
ages. FMRP expression in Iba-1 and NG2-positive cells was detected at postnatal day 0 and 10 and gradually decreased to very low or undetectable levels in postnatal day 20 and adult mice. Our results reveal that in addition to continuous and extensive expression in neurons in the immature and mature brain, FMRP is also present in astrocytes, oligodendrocyte precursor cells, and microglia during the early and mid-postnatal developmental stages of brain maturation. Prominent expression of FMRP in glia during these crucial stages of brain development suggests an important contribution to normal brain function, and in its absence, to the fragile X phenotype. & 2014 Elsevier B.V. All rights reserved.
1.
Introduction
of the FMR1 gene which induces a dramatic reduction or elimination of the expression of the encoded protein, fragile X
Fragile X syndrome (FXS) is a genetic disorder and a leading
mental retardation protein (FMRP). FMRP is an mRNA binding
cause of cognitive impairment and autism. The disorder is
protein that controls the expression of hundreds of genes in the
caused by triplet repeat expansion in the 5’ untranslated region
CNS through multiple mechanisms including modulating
n Correspondence to: Department of Pharmaceutical Sciences, University of Toronto, 144 College St. Toronto, Ontario, Canada M5S 3M2. E-mail address: [email protected] (D. Hampson).
http://dx.doi.org/10.1016/j.brainres.2014.11.023 0006-8993/& 2014 Elsevier B.V. All rights reserved.
23
brain research 1596 (2015) 22–30
ribosome stalling (Darnell et al., 2011). However, FMRP has been shown to also act as a positive modular of protein translation (Bechara et al., 2009), likely by enhancing mRNA stability (Zalfa et al., 2007). The FMR1 gene undergoes alternative splicing and possesses alternative transcription start sites such that at least 12 isoforms are generated (Brackett et al., 2013; Tassone et al., 2011). Isoform 1 is the longest form and codes for a protein of 632 amino acids and a molecular weight of 71kilodaltons. In the adult brain FMRP is abundantly expressed in most neurons throughout the CNS. In neurons, the protein is primarily located in the cytosol and in synaptic spines where it plays a role in spine maturation (Cruz-Martín et al., 2010). A portion of total cellular FMRP is also present in the nucleus where its role is less well characterized, and nuclear expression is isoform dependent (Dury et al., 2013). FMRP is highly abundant in “fragile X granules” in neuronal axons and presynaptic terminals where it apparently regulates recurrent neuronal activity (Akins et al., 2012). FMRP is also present in neural stem cells (Luo et al., 2010) where it has been shown to control hippocampal-dependent learning in the mature brain (Guo et al., 2011). In contrast to the expression of FMRP in neurons of the adult brain, relatively little is known about the types of glial cells that express FMRP during CNS development. FMRP was reported in primary cultures of rodent astrocytes (Yuskaitis et al., 2010), and in the mouse hippocampus, FMRP is expressed in astrocytes within the first week of birth and then declines to low or undetectable levels (Pacey and Doering, 2007). It is also present in oligodendroglia (Wang et al., 2004) and oligodendrocyte precursor cells (OPCs) in the immature cerebellum where it appears to be a factor in the proper progression of myelination (Pacey et al., 2013). The purpose of the present study was to more fully document the cell type expression of FMRP in neurons and glia in selected brain regions of the developing and mature mouse CNS. Our analysis encompassed the cerebral cortex, striatum, hippocampus, cerebellum, and the corpus callosum, and extended from postnatal days (PNDs) 0, to PNDs 10, 20, and adult. Our findings reveal that FMRP is extensively expressed in glia in the developing postnatal brain and that with the exception of the corpus callosum, glial expression declines at different rates in different brain regions to low levels in the adult brain.
2.
Results
Immunocytochemical analysis was used to map the expression of FMRP in selected brain regions at PND 0 (day of birth), 10, 20, and young adult (8–12 weeks old) wild-type C57/BL6 mice. Sections from cingulate cortex, hippocampus, striatum, corpus callosum, and cerebellum were immunolabeled for FMRP as well as a cell-type specific marker for neurons, astrocytes, oligodendrocyte precursor cells, and microglia. The 5C2 monoclonal antibody used here has previously been shown to be specific for FMRP (LaFauci et al., 2013). Brain sections treated with secondary antibody only showed no fluorescence. We quantified the total number of cells expressing FMRP per visual field, as well as cells co-expressing FMRP with one of the four cell-specific markers, and calculated the percent of total FMRPpositive cells that co-expressed each of the four cell markers in the cingulate cortex and corpus callosum (Table 1).
2.1. Gradual reduction in the number of FMRP-positive cells from the early postnatal period to adulthood FMRPþ cell counts in the cingulate cortex and corpus callosum revealed that the number of FMRPþ cells in both brain regions was highest at PND 0 (see “Total FMRPþ” column in Table 1). In the cingulate cortex there was a gradual decline over time such that the adult level was approximately 57% of that at birth (P0). In the corpus callosum there was a modest decline whereby adult mice showed about 85% of the level at P0. Overall, these observations are consistent with previous reports suggestion developmentally declining levels of FMRP protein, peaking at first postnatal week and declining thereafter (Lu et al., 2004; Pacey et al., 2013).
2.2. Predominant neuronal expression of FMRP throughout development At all postnatal times examined, FMRP expression was predominantly neuronal in all of the brain regions analyzed, except for corpus callosum (Fig. 1, Tables 1 and 2) where only a few NeuNþ cells were present and most FMRP expression was observed in S100βþ cells (Fig. 2, Table 2). Quantitative analysis of cell type-specific expression of FMRP revealed
Table 1 – Quantitative analysis of cell type-specific expression of FMRP in the cingulate cortex and corpus callosum. The analysis was performed by counting the FMRP-positive neurons (NeuN), astrocytes (S100β), microglia (Iba-1) and oligodendrocyte precursor cells (NG2) in the cingulate cortex and corpus callosum. The results are reported as the average number of total FMRP-positive cells per visual field, and the percentage of FMRP-positive cells that co-localized with NeuN, S100β, Iba-1, and NG2. Data are presented as mean7S.E.M. Standard error values smaller than 1.0 are not listed. Brain region
Age
Total FMRPþ
% NeuNþ
% S100β
% Iba-1
% NG2
Cingulate cortex
PND 0 PND 10 PND 20 Adult
15475 12375 8675 8877
85 8971 9471 9471
5 2 2 3
4 5 2 2
6 3 4 571
Corpus callosum
PND 0 PND 10 PND 20 Adult
8273 7573 7673 7073
2 2 2 2
27 3373 3372 2273
20 1973 1071 3
18 14 11 7
24
brain research 1596 (2015) 22–30
Fig. 1 – Expression of FMRP in neurons of the wild-type mouse brain. Coronal sections from postnatal days 0, 10, 20, and adult mice were double immunolabeled using anti-FMRP and anti-NeuN, and counterstained with DAPI (A–T). Co-localization of FMRP in NeuN-positive neurons (FMRP, green; NeuN, red; DAPI, blue) is shown in the cingulate cortex, hippocampus, striatum, corpus callosum, and cerebellum. Arrows indicate examples of co-localization. The heavy dashed lines delineate the boundaries of the corpus callosum. Scale bar¼20 μm.
85%–94% co-expression with NeuN in FMRPþ cells in the cingulate cortex. In stark contrast only 2% NeuN/FMRP coexpression was detected in the corpus callosum at all of the time points studied.
2.3. Developmental expression of FMRP in astrocytes, microglia and oligodendrocyte precursor cells We examined FMRP expression in astrocytes by double labeling with anti-FMRP and anti-S100β. FMRP and S100β were co-localized at PND 0 in cells in the cingulate cortex, striatum, hippocampus, corpus callosum, and cerebellum (Fig. 2, Tables 1 and 2). By PND 10, PND 20, and adult mouse brain, the population of FMRP/S100β -positive cells had declined drastically in cingulate cortex and hippocampus, (less than 3%). Compared to the cingulate cortex and hippocampus, the decline in FMRP/S100 β co-expression occurred more slowly in the striatum where substantial co-expression was seen at PND 10, PND 20, but not in adult mice. Persistent coexpression of FMRP/S100β was also observed in the cerebellum at PND0, 10, 20 and adult mice, although the number of FMRP/S100β cell gradually decreased in adult mouse brain. Interestingly, FMRP was co-expressed in many S100βþ cells in the corpus callosum. Of the brain regions analyzed, the corpus callosum showed by far the largest percentage of cells co-expressing of FMRP and S100β (Fig. 2, Table 2); more than 75% of S100β positive cells in the corpus callosum co-expressed
FMRP at all time points. This is an intriguing finding as this brain region is mainly composed of white matter. Approximately 30% of FMRP-expressing cells in the corpus callosum also expressed S100β at all developmental stages analyzed (Tables 1 and 2). The percent of S100β/FMRP-positive cells was maintained at a moderate level from PND 0 to adult mice, suggesting a continual expression of FMRP in astrocytes from early postnatal stages through adulthood in this brain region. In addition to the expression of S100β in astrocytes, S100β is also expressed by a population of oligodendrocyte precursor cells and oligodendrocytes (Hachem et al., 2005; Steiner et al., 2008). Therefore, two population of S100β immunopositive cells may exist, one representing astrocytes and one dedicated to oligodendrocyte precursor cells. To better characterize these two populations of cells, we also used NG2 staining as a specific marker for oligodendrocyte precursor cells. The NG2 proteoglycan is expressed in oligodendrocyte precursor cells, but not in mature oligodendrocytes, both in vitro and in developing rodent brain (Polito and Reynolds, 2005). FMRP was expressed in many NG2þ oligodendrocyte precursor cells in the cingulate cortex, striatum, hippocampus, corpus callosum, and cerebellum at PND 0 (Fig. 3, Tables 1 and 2). Thereafter, from PND 10 to adult mouse brain, FMRP expression in NG2-positive cells progressively declined in all of the brain regions examined, reaching low co-localization in adult mice in the cingulate cortex,
25
brain research 1596 (2015) 22–30
Table 2 – Semi-quantitative analysis of cell type-specific FMRP expression in the developing and adult mouse brain. The level of co-localization of FMRP with NeuN-positive neurons, S100β-positive astrocytes, Iba-1-positive microglia, and NG2positive oligodendrocyte precursor cells in mouse brain was assessed in the brain regions indicated at PND 0, PND 10, PND 20, and adults. The extent of co-localization is reported as: -, none;þlow; þþ moderate; and þþþ high. Brain region
Age of mice
FMRPþ/NeuNþ
FMRPþ/S100βþ
FMRPþ/Iba-1þ
FMRPþ/NG2þ
Cingulate cortex
PND 0 PND 10 PND 20 Adult
þþþ þþþ þþþ þþþ
þ þ þ þ
þ þ þ þ
þ þ þ þ
Corpus callosum
PND 0 PND 10 PND 20 Adult
þ þ þ þ
þþþ þþþ þþþ þþþ
þþþ þþþ þþ þ
þþþ þþ þþ þ
Striatum
PND 0 PND 10 PND 20 Adult
þþþ þþþ þþþ þþþ
þþ þ þ þ
þþ þ þ -
þþ þ þ þ
Hippocampus
PND 0 PND 10 PND 20 Adult
þþþ þþþ þþþ þþþ
þþ þ -
þþ þ -
þ þ þ -
Cerebellum
PND 0 PND 10 PND 20 Adult
þþþ þþþ þþþ þþþ
þ þ þ þ
þ þ -
þ þþ þ þ
Fig. 2 – Expression of FMRP in the astrocytes of mouse brain. Coronal sections from postnatal days 0, 10, 20, and adult mice were labeled using anti-FMRP, anti-S100β, and DAPI. FMRP in S100β -positive astrocytes (FMRP, green; S100β, red; DAPI, blue) are shown in the cingulate cortex, hippocampus, striatum, corpus callosum, and cerebellum. Arrows indicate examples of colocalization of FMRP in S100β-positive astrocytes, whereas dashed-squares indicate the absence of FMRP in astrocytes. The heavy dashed lines delineate the boundaries of the corpus callosum. Scale bar¼ 20 μm.
26
brain research 1596 (2015) 22–30
Fig. 3 – Expression of FMRP in oligodendrocyte precursor cells in the mouse brain. Coronal sections were labeled using antiFMRP, anti-NG2, and DAPI. Co-localization of endogenous FMRP in NG2-positive oligodendrocyte precursor cells (FMRP, green; NG2, red; DAPI, blue) is shown in the cingulate cortex, hippocampus, striatum, corpus callosum, and cerebellum. Arrows indicate the co-localization of FMRP in NG2-positive oligodendrocyte precursor cells, whereas dashed-squares indicate the absence of FMRP in oligodendrocyte precursor cells. The heavy dashed lines delineate the boundaries of the corpus callosum. Scale bar¼ 20 μm.
Fig. 4 – Expression of FMRP in microglia. Coronal sections were labeled with anti-FMRP and anti-Iba-1. Co-localization of FMRP in Iba-1-positive microglia (FMRP, green; Iba1, red; DAPI, blue) is shown. Arrows indicate examples of co-localization of FMRP in Iba-1-positive microglia, whereas dashed-squares indicate the absence of FMRP in microglia. Scale bar¼20 μm.
brain research 1596 (2015) 22–30
corpus callosum, striatum, and cerebellum and undetectable co-localization in the hippocampus (Table 2). FMRP expression in microglia was examined by assessing co-expression with Iba-1. FMRP/ Iba-1 double labeled cells were present in all brain regions analyzed at PND 0 (Fig. 4). At PND 0, more than 75% of the total NG2þ cells and Iba-1þ cells co-expressed FMRP in the all the brain regions analyzed. At PND 10 and PND 20, the number of FMRP/Iba-1 co-localized cells remained relatively constant in the cingulate cortex, corpus callosum, and striatum, but decreased in the hippocampus and cerebellum by PND 20. In the brain regions studied here, there was a gradual decrease in FMRP/Iba-1 co-expression after PND 20 with no detectable co-localization in adult mice.
3.
Discussion
Our results demonstrate a predominantly neuronal expression of FMRP in all of the brain regions analyzed, except the corpus callosum. Our data also reveal the presence of FMRP in cells expressing markers for astrocytes (S100β), microglia (Iba1) and oligodendrocyte precursor cells (NG2) in the developing brain. In general, glial expression of FMRP diminished to low or undetectable levels in the adult brain following different trajectories in different brain regions. FMRP expression was abundant in astrocytes at PND 0 in all of the brain regions analyzed; thereafter, expression in astrocytes showed a gradual decline in striatum and hippocampus, reaching negligible numbers of FMRPþ/S100βþ cells in the adult mouse brain. Pacey and Doering (2007) reported FMRP expression in cells expressing the astrocytic marker GFAP in the hippocampus and the ependymal cells surrounding the third ventricle at PND 1 and PND 7, but no FMRP/GFAP co-localization in the brains of adult (2 month-old) mice. Our results confirmed the findings of Pacey and Doering, (2007) in the hippocampus using a different astrocytic marker (S100β), and extended these findings to other brain regions including the cingulate cortex, corpus callosum, striatum, and cerebellum. Quantitative analysis of cell-type expression of FMRP indicated that more than 20% of FMRPþ cells in the corpus callosum co-expressed S100β and this co-localization rate remained relatively constant from PND 0 to adult mice. In contrast, there were few NeuNþ cells detected in the corpus callosum. Expression of FMRP in astrocytes could be of crucial importance in regulating the established functions of astrocytes in synapse formation and neural plasticity. More specifically, during the early postnatal weeks in rodents when astrocyte support of neuron growth and synapse formation is vital, the lack of FMRP could contribute to the abnormal dendritic morphology and synapse development seen in FXS (Jacobs and Doering, 2010). Astrocytes dynamically regulate synaptic transmission, partly by releasing soluble synaptogenic molecules such as the astrocyte-secreted matricellular proteins, including Hevin, and thrombospondin-1 (Jones and Bouvier, 2014, or neurotrophic factors such as Neurotrophin-3 (Yang et al., 2012), which are highly expressed within first few weeks of postnatal development and decrease as the brain matures. Interestingly, Hevin, also known as SPARC-like 1, and Neurotrophin-3 have been identified as an mRNA
27
substrates for FMRP (Darnell et al., 2011) and excessive neurotrophin-3, secreted from Fmr1 knockout astrocytes, has been suggested to contribute to abnormal neuronal dendritic development in Fmr1 knockout mouse model (Yang et al., 2012). Furthermore, thrombospondin-1 protein is reduced in the astrocytes of Fmr1 knockout mice and synaptic deficits are corrected by its protein replacement in Fmr1 knockout mouse hippocampal neurons (Cheng et al., 2014). These findings highlight the important role for FMRP expression in astrocytes during early postnatal weeks, which coincide with the peak of synaptogenesis. FMRP expression has been reported in oligodendrocyte precursors in vitro (Wang et al., 2004) and in the early postnatal cerebellum (Pacey et al., 2013). Our data extend these findings by reporting a more comprehensive analysis of FMRP/NG2 co-localization in different brain regions at multiple time points during postnatal development of the CNS. FMRP expression is also present in mature MBPþ oligodendrocytes of rodents as well as human oligodendrocytes (Giampetruzzi et al., 2013). FMRP was previously shown to bind to MBP mRNA and down-regulate its translation in vitro (Wang et al., 2004). Moreover, the lack of FMRP is associated with delayed myelination in the Fmr1 knockout mouse cerebellum (Pacey et al., 2013). In the present study, we observed a gradual decline in the number of FMRPþ cells co-expressing NG2 from PND 0 to adult mice; this pattern was observed in all of the brain regions examined, but was most prominent in the corpus callosum and striatum. A decline in FMRP levels during oligodendrocyte differentiation was previously reported in primary cultures of oligodendrocytes derived from neonatal rat brain (Wang et al., 2004). This gradual decline indicates that FMRP likely plays an important role in myelin formation during CNS maturation. FMRP has been previously reported in cultures of microglial cells (Yuskaitis et al., 2009; Yuskaitis et al., 2010). However, the findings reported here provide the first direct evidence for FMRP expression in microglia in brain tissue. Post mortem studies have identified neuroinflammation and glial activation in the brains of individuals with idiopathic autism (Fatemi et al., 2011; Tetreault et al., 2012), and other studies suggest a similar pathology may be present in fragile X syndrome (Jacobs et al., 2012). FMRP expression was reported in the BV2 microglial cell line (Yuskaitis and Jope, 2009). However, despite microglial expression of FMRP, there was no difference in the production of pro-inflammatory cytokines IL-6 and TNFα after acute lipopolysaccharide induced activation of microglia in wild-type or Fmr1 knockout mice. In the current study, FMRP expression in microglia was detected most prominently during the first 10 postnatal days and declined thereafter. Interestingly, the number of microglial cells in mice is immensely increased during the first two postnatal weeks when the vast majority (495%) of microglia are born (Alliot et al., 1999). These findings suggest a potential regulatory role for FMRP in the cell lineage and maturation of microglia during early development. However, the exact molecular function of FMRP in these cells and its contribution to behavioral abnormalities in Fmr1 knockout mice remains to be elucidated. Quantitative analysis of FMRPþ cells in the cingulate cortex and corpus callosum demonstrated the highest number of FMRPþ cells at PND 0 which diminished steadily throughout
28
brain research 1596 (2015) 22–30
brain development. The abundant FMRP expression at birth and during the first 1–2 postnatal weeks indicates that the functional requirement of FMRP in glia is highest during this critical period of early brain development. In the drosophila nervous system FMRP appears to be present sequentially, first in neuroblasts and then in glia, and glial expression was required for neuroblast reactivation (Callan et al., 2012). Similar to the mouse model, the expression of the drosophila FMRP homolog is developmentally regulated (Tessier et al., 2008) and its reintroduction during brain development rescues the dendritic impairment in the drosophila model (Gatto et al., 2009), indicating a causal relationship between FMRP expression during a distinct developmental time-window and normal synapse formation. In the mouse CNS, the most intense period of synaptogenesis (Zito and Svoboda, 2002) and the most severe dendritic spine abnormality in Fmr1 knockout brain were found to coincide at the end of the first postnatal week when abundant FMRP expression occurs (Nimchinsky et al., 2001). It has been shown that FMRP re-introduction specifically in CNS neurons using viral vector-mediated gene therapy was able to correct selected behavioral abnormalities in the Fmr1 knockout mouse (Gholizadeh et al., 2014). Conceivably, the reintroduction of FMRP in neurons could indirectly result in correction of disrupted neuron-to-astrocyte signaling in FXS at a critical time during early development when astrocyte support of neuronal growth and synapse formation is vital (Cheng, et al., 2012). Since glia play indispensable roles in synaptic development (Corty and Freeman, 2013), developmentally regulated FMRP expression in glia may influence synaptic development of adjacent neurons. Future efforts could be directed toward determining whether FMRP reintroduction in astrocytes alone, or concomitantly in both astrocytes and neurons, may provide enhanced benefit (correction of abnormal behaviors) over that seen when expressed in neurons only (Glolizadeh et al., 2014).
4.
Experimental procedures
4.1.
Animals
Wild-type C57BL/6 J mice were bred at the University of Toronto and used for all the experiments. Mice were kept in a room maintained at constant temperature (2172 1C) and relative humidity (5575%) with an automatic 12 h light/dark cycle with free access to standard laboratory diet and tap water. All animal experiments were carried out in accordance with the guidelines set out by the Canadian Council on Animal Care and were approved by the University of Toronto Animal Care Committee.
4.2.
Tissue collection and preparation
PND 10, PND 20, and adult (PND 55–65) wild-type mice were anaesthetized with a mixture of ketamine and xylazine (intraperitoneal, 80 and 8 mg/kg, respectively) and perfused transcardially with a solution of 0.1 M phosphate buffer saline (PBS, pH 7.4), followed by 4% paraformaldehyde (PFA, pH 7.4). The brains were removed and post-fixed overnight in 4% PFA at 4oC, and then transferred to 30% sucrose solution for cryoprotection. For PND 0, pups were decapitated, rinsed
with PBS and the skull was fixed in 4% PFA at room temperature for 2–4 h. The brain was then removed from the skull and further incubated in 4% PFA at 4oC for 20–24 h followed by 30% sucrose solution overnight at 4 1C. Once the brains sank in the sucrose, they were mounted in optimum cutting temperature solution (Sakura, Torrance, CA) and frozen at -80 1C until sectioning. For PND 10, PND 20 and adult mice, floating coronal brain sections were prepared at a thickness of 25 mm using a cryostat (Leica Microsystems, Wetzlar, Germany). For PND 0, both floating and thawmount sections were used (25 μm thick).
4.3.
Immunohistochemistry and confocal microscopy
To perform double fluorescence immunostaining, coronal brain sections were washed three times with Tris-buffered saline (TBS, pH 7.6) for 5 min each at room temperature (RT). The antigen retrieval assay was performed at all times in which sections were incubated with 0.8% sodium borohydride for 10 min at RT followed by washing once with TBS, and then incubated with 0.01 M citrate buffer (pH 6.0) in the water bath at 75 1C for 45 min. After cooling for 30 min, sections were washed three times with TBS and blocked for 1 h at RT with TBS containing and 0.1% Triton X-100, 5% goat serum (SigmaAldrich, St. Louis, MO, USA) and goat anti-mouse IgG (dilution 1:50; Sigma-Aldrich). Sections were washed four times in TBS for 5 min each with gentle rocking and then incubated overnight at 4 1C in the following primary antibodies diluted in TBS containing 5% goat serum: mouse monoclonal anti-FMRP 5c2 (1:1000) (LaFauci et al., 2013), rabbit polyclonal anti-NeuN (1:2000; Cat. # ab128886, Abcam, Cambridge, UK), rabbit polyclonal anti-S100β (1:1000; Cat. # ab868, Abcam, Cambridge, UK), rabbit polyclonal anti-Iba-1 (1:2000; Cat. # 019-19741, WAKO, Richmond, VA, USA), and rabbit polyclonal anti-NG2 (1:500; Cat. # AB5320, Millipore, Temecula, CA, USA). Sections were washed with TBS five times for 10 min each and then incubated with following secondary antibodies (diluted in TBS containing 5% goat serum) for 2 h at RT in the dark with gentle shaking: goat anti-mouse Alexa Fluor 488 and goat antirabbit Alexa Fluor 594 (diluted 1:1000; Jackson ImmunoResearch Laboratories, West Grove, PA, USA). The nuclei were visualized with DAPI (1:80000; Sigma-Aldrich). Sections were then washed in PBS (five times, 10 min each) with minimal exposure to light, mounted on glass slides, air dried, and then coverslipped with Prolong Gold antifade mounting medium (Invitrogen, Carlsbad, CA, USA). Representative tissue sections covering discrete brain regions including cingulate cortex, corpus callosum, striatum, hippocampus, and cerebellum were examined with a laser-scanning confocal microscope (Nikon A1; Nikon Instruments, Tokyo, Japan). The images were captured at 60X magnification and analyzed with Image J software (NIH, Bethesda, MD, USA) and NIS-Elements software (Nikon Instruments, Tokyo, Japan).
4.4.
Semi-quantitative and quantitative analysis
For comparative analysis of cell type-specific FMRP expression in the PND 0, PND 10, PND 20 and adult mouse brain, two scoring methods were used: a quantitative scoring system to count the FMRP-positive neurons (NeuN as marker), astrocytes (S100β as
brain research 1596 (2015) 22–30
marker), microglia (Iba-1 as marker) and oligodendrocyte precursor cells (NG2 as marker) in the cingulate cortex and corpus callosum, and a semi-quantitative scoring system to analyze the FMRP co-localization with NeuN, S100β, Iba-1, and NG2 in neurons, astrocytes, microglia, and oligodendrocyte precursor cells in the different brain regions including cingulate cortex, corpus callosum, striatum, hippocampus, and cerebellum. Briefly, photomicrographs made with the 60X objective lens were used to assess the number of FMRP-positive neurons, astrocytes, microglia, and oligodendrocyte precursor cells in the different brain regions. The number of cells in each bright field image was counted using specific cell markers with clearly visible nuclei by DAPI staining in random square fields (45000 mm 45000 μm) in the coronal brain sections (Table 2). Cell counting was performed using three consecutive coronal brain sections from each PND 0, PND 10, PND 20, and adult mouse (total of 9 sections from three mice at each age group) and reported as the average number of FMRP-positive cells per visual field, and the average percentage of FMRP-positive cells that co-localized with NeuN, S100β, Iba-1, and NG2 to obtain a mean value for each mouse. For semi-quantitative analysis, a scoring system was used to analyze the co-localization in FMRPþ cells in the cingulate cortex, hippocampus, striatum, corpus callosum, and cerebellum. Levels of co-localization are presented as the percentage of FMRPþ cells that co-localized with each of the cell-specific markers (shown in Table 2). The results were summarized as: -, no co-localization; þ, low co-localization; þþ, moderate co-localization; and þþþ, high co-localization.
Acknowledgments We thank Dr. Laura K.K. Pacey for helpful comments on the manuscript. This work was supported by an operating grant from the Canadian Institutes of Health Research and the Canadian Institutes of Health Research Training Program in Biological Therapeutics.
r e f e r e n c e s
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. Alliot, F., Godin, I., Pessac, B., 1999. Microglia derive from progenitors, originating from the yolk sac, and which proliferate in the brain. Brain Res. Dev. Brain Res. 117, 145–152. Bechara, E., Didiot, M., Melko, M., Davidovic, L., Bensaid, M., Martin, P., Castets, M., Pognonec, P., Khandjian, E., Moine, H., Bardoni, B., 2009. A novel function for fragile X mental retardation protein in translational activation. PLoS Biol 7, e16. Brackett, D.M., Qing, F., Amieux, P.S., Sellers, D.L., Horner, P.J., Morris, D.R., 2013. FMR1 transcript isoforms: association with polyribosomes; regional and developmental expression in mouse brain. PLoS One 8, e58296. Callan, M.A., Clements, N., Ahrendt, N., Zarnescu, D.C., 2012. Fragile X Protein is required for inhibition of insulin signaling and regulates glial-dependent neuroblast reactivation in the developing brain. Brain Res. 1462, 151–161. Cheng, C., Sourial, M., Doering, L.C., 2012. Astrocytes and developmental plasticity in fragile X. Neural Plast 2012, 197491.
29
Cheng, C., Zhu, L., Doering, L.C., 2014. Astrocyte-secreted TSP-1 corrects synaptic and spine phenotypes in the fragile X mouse model. Fourteenth International Fragile X Conference., Orange County, CA. Corty, M.M., Freeman, M.R., 2013. Cell biology in neuroscience: architects in neural circuit design: glia control neuron numbers and connectivity. J. Cell Biol. 203, 395–405. Cruz-Martı´n, A., Crespo, M., Portera-Cailliau, C., 2010. Delayed stabilization of dendritic spines in fragile X mice. J. Neurosci. 30, 7793–7803. Darnell, J., Van Driesche, S., Zhang, C., Hung, K., Mele, A., Fraser, C., Stone, E., Chen, C., Fak, J., Chi, S., Licatalosi, D., Richter, J., Darnell, R., 2011. FMRP stalls ribosomal translocation on mRNAs linked to synaptic function and autism. Cell 146, 247–261. Dury, A.Y., El Fatimy, R., Tremblay, S., Rose, T.M., Cote, J., De Koninck, P., Khandjian, E.W., 2013. Nuclear fragile X mental retardation protein is localized to Cajal bodies. PLoS Genet. 9, e1003890. Gholizadeh, S., Arsenault, J., Xuan, I.C., Pacey, L.K., Hampson, D.R., 2014. Reduced phenotypic severity following adenoassociated virus-mediated Fmr1 gene delivery in fragile X mice. Neuropsychopharmacology. Giampetruzzi, A., Carson, J.H., Barbarese, E., 2013. FMRP and myelin protein expression in oligodendrocytes. Mol. Cell Neurosci. 56, 333–341. Guo, W., Allan, A., Zong, R., Zhang, L., Johnson, E., Schaller, E., Murthy, A., Goggin, S., Eisch, A., Oostra, B., Nelson, D., Jin, P., Zhao, X., 2011. Ablation of Fmrp in adult neural stem cells disrupts hippocampus-dependent learning. Nat. Med 17, 559–565. Hachem, S., Aguirre, A., Vives, V., Marks, A., Gallo, V., Legraverend, C., 2005. Spatial and temporal expression of S100B in cells of oligodendrocyte lineage. Glia 51, 81–97. Jacobs, S., Doering, L.C., 2010. Astrocytes prevent abnormal neuronal development in the fragile x mouse. J. Neurosci. 30, 4508–4514. Jacobs, S., Cheng, C., Doering, L.C., 2012. Probing astrocyte function in fragile X syndrome. Results Probl. Cell Differ. 54, 15–31. Jones, E.V., Bouvier, D.S., 2014. Astrocyte-secreted matricellular proteins in CNS remodelling during development and disease. Neural Plast 2014, 321209. LaFauci, G., Adayev, T., Kascsak, R., Nolin, S., Mehta, P., Brown, W.T., Dobkin, C., 2013. Fragile X screening by quantification of FMRP in dried blood spots by a Luminex immunoassay. J. Mol. Diagn. 15, 508–517. Lu, R., Wang, H., Liang, Z., Ku, L., O’Donnell, W., Li, T., Warren, W., Feng, Y., S.T., 2004. The fragile X protein controls microtubuleassociated protein 1B translation and microtubule stability in brain neuron development. Proc. Natl. Acad. Sci. U. S. A 101, 15201–15206. Luo, Y., Shan, G., Guo, W., Smrt, R., Johnson, E., Li, X., Pfeiffer, R., Szulwach, K., Duan, R., Barkho, B., Li, W., Liu, C., Jin, P., Zhao, X., 2010. Fragile x mental retardation protein regulates proliferation and differentiation of adult neural stem/ progenitor cells. PLoS Genet 6, e1000898. Nimchinsky, E., Oberlander, A., Svoboda, K., 2001. Abnormal development of dendritic spines in FMR1 knock-out mice. J. Neurosci. 21, 5139–5146. Pacey, L.K.K., Doering, L.C., 2007. Developmental expression of FMRP in the astrocyte lineage: implications for fragile X syndrome. Glia 55, 1601–1609. Pacey, L.K.K., Xuan, I.C., Guan, S., Sussman, D., Henkelman, R.M., Chen, Y., Thomsen, C., Hampson, D.R., 2013. Delayed myelination in a mouse model of fragile X syndrome. Hum. Mol. Genet. 22, 3920–3930.
30
brain research 1596 (2015) 22–30
Polito, A., Reynolds, R., 2005. NG2-expressing cells as oligodendrocyte progenitors in the normal and demyelinated adult central nervous system. J. Anat. 207, 707–716. Steiner, J., Bernstein, H.G., Bogerts, B., Gos, T., Richter-Landsberg, C., Wunderlich, M.T., Keilhoff, G., 2008. S100B is expressed in, and released from, OLN-93 oligodendrocytes: Influence of serum and glucose deprivation. Neuroscience 154, 496–503. Tassone, F., De Rubeis, S., Carosi, C., La Fata, G., Serpa, G., Raske, C., Willemsen, R., Hagerman, P.J., Bagni, C., 2011. Differential usage of transcriptional start sites and polyadenylation sites in FMR1 premutation alleles. Nucleic Acids Res. 39, 6172–6185. Tetreault, N.A., Hakeem, A.Y., Jiang, S., Williams, B.A., Allman, E., Wold, B.J., Allman, J.M., 2012. Microglia in the cerebral cortex in autism. J. Autism. Dev. Disord. 42, 2569–2584. Wang, H., Ku, L., Osterhout, D., Li, W., Ahmadian, A., Liang, Z., Feng, Y., 2004. Developmentally-programmed FMRP expression in oligodendrocytes: a potential role of FMRP in regulating translation in oligodendroglia progenitors. Hum. Mol. Genet. 13, 79–89.
Yang, Q., Feng, B., Zhang, K., Guo, Y.Y., Liu, S.B., Wu, Y.M., Li, X.Q., Zhao, M.G., 2012. Excessive astrocyte-derived neurotrophin-3 contributes to the abnormal neuronal dendritic development in a mouse model of fragile X syndrome. PLoS Genet. 8, e1003172. Yuskaitis, C.J., Jope, R.S., 2009. Glycogen synthase kinase-3 regulates microglial migration, inflammation, and inflammation-induced neurotoxicity. Cell Signal. 21, 264–273. Yuskaitis, C.J., Beurel, E., Jope, R.S., 2010. Evidence of reactive astrocytes but not peripheral immune system activation in a mouse model of Fragile X syndrome. Biochim. Biophys. Acta. 1802, 1006–1012. Zalfa, F., Eleuteri, B., Dickson, K., Mercaldo, V., De Rubeis, S., di Penta, A., Tabolacci, E., Chiurazzi, P., Neri, G., Grant, S., Bagni, C., 2007. A new function for the fragile X mental retardation protein in regulation of PSD-95 mRNA stability. Nat. Neurosci. 10, 578–587. Zito, K., Svoboda, K., 2002. Activity-dependent synaptogenesis in the adult Mammalian cortex. Neuron 35, 1015–1017.
