|Home | About | Journals | Submit | Contact Us | Français|
GPR56, one of the adhesion G-protein–coupled receptors (GPCRs), plays an important role in the development of the cerebral cortex. Mutations in GPR56 cause a severe human cortical malformation called bilateral frontoparietal polymicrogyria (BFPP), characterized by a global malformation of the cerebral cortex that most severely affects the frontal and parietal regions. To characterize the expression pattern of GPR56 in the developing cerebral cortex, we developed a mouse monoclonal antibody against mouse GPR56. We revealed that GPR56 is expressed in multiple cell types in the preplate, marginal zone, subventricular zone (SVZ), and ventricular zone (VZ). Most interestingly, the in preplate neurons showed an anterior-to-posterior gradient at embryonic day (E) 10.5–11.5. In contrast, the expression pattern of the GPR56 ligand, collagen III, revealed no visible gradient pattern. With the widespread expression of GPR56 in the developing cortex, it is difficult to draw a specific conclusion as to which of the GPR56-expressing cells are critical for human brain development. However, the correlation between GPR56 expression in neurons at E10.5–E11.5 and the anatomic distribution of the cortical malformation in both humans and mice suggests that its function in preplate neurons is indispensible.
Mutations in GPR56 cause bilateral frontoparietal polymicrogyria (BFPP), a devastating brain malformation (Piao et al., 2004, 2005). Individuals with BFPP present with mental retardation, language impairment, motor developmental delay, and seizure disorders (Chang et al., 2003; Piao et al., 2002, 2004, 2005). The observed cortical abnormality extends across the frontal and parietal lobes with a decreasing anterior-to-posterior gradient of severity. However, it remains unclear how the anatomic distribution of the malformation occurs.
Histological analysis of GPR56 knockout mouse brains showed that the loss of GPR56 results in breaches of the pial basement membrane (BM) and neuronal overmigration—a hallmark of cobblestone lissencephaly (Li et al., 2008). This histopathology was further confirmed by studies of postmortem human BFPP brains, which concretely demonstrated that BFPP is a cobblestone-like brain malformation (Bahi-Buisson et al., 2010). Although the leading pathology of cobblestone lissencephaly is thought to be the result of a regional breakdown of the pial BM, the most recent literature indicates that abnormal neuronal migration also accounts for the development of cobblestone lissencephaly (Jaglin et al., 2009; Kwiatkowski et al., 2007; Moers et al., 2008; Olson and Walsh, 2002; Sarkisian et al., 2006). We previously showed that radial glial cells express GPR56 in abundance in their endfeet, presumably regulating the integrity of the pial BM (Li et al., 2008).
However, our recent data demonstrate that the interaction of GPR56 and its ligand, collagen III, inhibits neuronal migration, revealing a yet uncharacterized function of GPR56 in cortical development (Luo et al., 2011). This current study aims to examine the expression pattern of GPR56 in the developing neocortex, thus laying the foundation for further study of the myriad of roles that GPR56 plays in cortical development.
The primary antibodies are described in detail in Table 1. The rabbit anti-Tuj1 monoclonal antibody (Covance, Princeton, NJ) was raised against microtubules derived from rat brains. This antibody detects a single band of 50 kDa on Western blot of rat brain cell extracts (manufacturer’s information). Our immunostaining pattern in postmitotic neurons of the mouse brain is identical to that in the previous report (Luo et al., 2011).
The rabbit anti-human collagen III antibody (Lifespan Biosciences Seattle, WA) was raised against full-length human type III collagen with an intact 3D helical structure. This antibody recognizes a single protein band between 160 and 260 kDa on Western blot of mouse meningeal fibroblasts (Luo et al., 2011). Our immunostaining of this antibody reveals a specific expression pattern, as previously reported (Luo et al., 2011).
The mouse anti-reelin antibody (Chemicon International, Billerica, MA) was raised against a recombinant reelin protein containing amino acids 164–496, with the epitope between amino acids 164 and 189. This antibody detects three protein bands; 420, 310, and 180 kDa on Western blot of cortical and cerebellar cell extract (Botella-Lopez et al., 2006). Our immunostaining pattern was identical to that in the previous report on the mouse cortex (Alcantara et al., 1998).
The rabbit anti-Tbr2 antibody (Abcam, Cambridge, MA) was raised against a synthetic peptide containing amino acids 650 to the C-terminus of mouse TBR2/Eomes conjugated to keyhole limpet hemocyanin. This antibody detects a clear band of 72 kDa on Western blot of cell lysates of EL4 cells expressing V5-tagged Eomesodermin. However, it shows two bands of ~85 and ~75 kDa of either human mesendoderm or E14 mouse embryonic brains (manufacturer’s information). Our immunostaining of this antibody showed the same distribution of Tbr2 transcripts in the mouse neocortex as that previously reported (Bulfone et al., 1999).
The calretinin antibody (Swant, Bellinzona, Switzerland) was raised against a recombinant human calretinin containing a 6-His tag at the N-terminus. This antibody recognizes a single protein band of 29–30 kDa on Western blot of brain homogenates of various species including mouse, rat, guinea pig, rabbit, Macaca fascicularis, zebrafish, and chicken (manufacturer’s information). The antibody specificity was confirmed by immunohistochemistry (IHC) on calretinin knockout mouse brain (manufacturer’s information). Our immunostaining of this antibody showed the same distribution of calretinin in the mouse neocortex as previously reported (Hevner et al., 2003).
Timed pregnant wild-type mice of CD-1 background were ordered from Charles River (Wilmington, MA). The Gpr56 knockout mice were kindly provided by Genentech (South San Francisco, CA) and produced in collaboration between Genentech and Lexicon Genetics (Woodlands, TX) to analyze the function of about 500 secreted and transmembrane proteins. All animals were treated according to the guidelines of the Animal Care and Use Committee of Children’s Hospital Boston.
Histological analysis was carried out as previously described with minor modifications (Brown et al., 1996; Li et al., 2008). Embryonic brains were fixed in 4% paraformaldehyde at 4°C for either 3 hours (E10.5, E11.5, and E12.5) or overnight (E14.5 and E18.5). After protecting in 30% sucrose and embedding in OCT, frozen sections were prepared at 8 μm on a cryostat (Leica, Deerfield, IL). For H11 antibody immunostaining, antigens were retrieved by boiling the slides in Retrievagen A Solution (BD Pharmingen, San Diego, CA) for either 8 minutes (E10.5, E11.5, and E12.5), 12 minutes (E14.5), or 14 minutes (E18.5), followed by cooling down at room temperature for 30 minutes. The slides were then washed three times in phosphate-buffered saline (PBS), once in 1% sodium dodecyl sulfate (SDS) in PBS, and then again three times in PBS. After blocking with 10% goat serum, 1% bovine serum albumin (BSA), and 0.1% Triton X-100 in PBS for 1 hour at room temperature, the sections were incubated with antibody H11 overnight at 4°C followed by washing once in PBS, twice in 2.7% NaCl PBS (instead of 0.8% NaCl), and once more in PBS. Primary antibodies were visualized by incubation in Alexa Fluor 488 or Alexa Fluor 546 (1:1,000; Invitrogen, La Jolla, CA) for 30 minutes at room temperature. Counterstaining was carried out with Hoechst 33342 (1:2,000; Invitrogen).
Primary neural culture was performed as previously described (Luo et al., 2011). Briefly, cortical cells were harvested from E13.5 mouse cortices with the meninges removed. Dissociated cells were seeded on 10-ml tissue culture dishes at 37°C for 10 minutes to deplete the fibroblasts. The cells (1 × 105/ml) were placed on cover-slips precoated with poly-D-lysine (100 μg/ml) and cultured in neural culture medium (neurobasal medium supplemented with B27, 1% penicillin/streptomycin, and 1% L-glutamate) for 1 day.
To perform immunocytochemistry, the cultured primary neuronal cells were fixed in cold acetone for antibody 199 or cold 95% ethanol and 5% glacial acetic acid for antibody H11, followed by three washes with PBS. Cells were permeabilized with 0.1% Triton-X 100 in PBS for 10 minutes followed by three washes with PBS. After blocking with 10% goat serum, 1% BSA, and 0.1% Triton-X100 in PBS for 30 minutes, the primary antibodies were incubated at 4°C overnight and visualized by appropriate fluorophore-conjugated secondary antibodies (1:1,000; Invitrogen). Nuclei were stained with Hoechst 33342 (1:2,000; Invitrogen). Images were captured by using a Nikon 80i upright microscope.
Sagittal sections of E10.5–E12.5 brains (three different brains for each developmental stage) were stained with antibody H11 as described above. The images were captured by using a Nikon 80i upright microscope. The selected region of the preplate and VZ was evenly divided into 10 bins. Fluorescent density in each bin was measured by using Image QuantTL (Amersham Bioscience, Arlington Heights, IL). The fluorescent density was subtracted from the background and divided by the density of the first bin to obtain the relative fluorescent intensity. Two-tailed Student’s t-tests were performed for P values.
The mouse monoclonal antibody was raised against the N-terminal fragment of mouse GPR56 (GPR56N) at the Dana Farber/Harvard Cancer Center Monoclonal Antibody Core. Naturally glycosylated GPR56N-mFc fusion protein (containing amino acids 27–382) was purified from the conditioned media of transfected HEK293T cells and used to immunize Gpr56 knockout mice to generate hybridoma cells. One positive clone, H11, was selected based on its high specificity on both Western blot analysis and IHC (Fig. 1).
We have previously shown that radial glial cells express GPR56 (Li et al., 2008). However, in an effort to characterize the ligand of GPR56, we recently discovered that the interaction of GPR56 and its ligand, collagen III, inhibits neuronal migration, suggesting a role for GPR56 in migrating neurons (Luo et al., 2011). To examine whether GPR56 is expressed by postmitotic neurons, we performed double IHC of GPR56 and Tuj1 in mouse embryonic brains at E12.5, E14.5, and E18.5. We observed a co-expression of GPR56 and Tuj1 in the pre-plate and marginal zone/subpial neurons in E12.5 and E14.5 brains (Fig. 2A–F). However, GPR56 expression was largely restricted to the ventricular zone (VZ) at E18.5 (Fig. 2J–L). To further demonstrate that Tuj1-positive neurons truly express GPR56, we performed a double immunostaining using H11 and Tuj1 antibodies on 1 day in vitro (1DIV) cultured progenitor cells (Fig. 2P–R). As shown in Figure 2R, some Tuj1-positive cells indeed express GPR56.
To explore the etiology of the gradient cortical phenotype associated with GPR56 mutations, we first examined the expression pattern of collagen III, the ligand of GPR56. Double IHC of collagen III and GPR56 was conducted on sagittal sections of mouse embryonic brains ranging in age from E10.5 to E11.5. From this analysis, we found that collagen III is largely expressed in the meninges and pial BM, yet no visible expression gradient of collagen III was detected during these developmental stages (Fig. 3A,D).
Interestingly, we observed an anterior-to-posterior gradient expression of GPR56 in the preplate neurons in both E10.5 and E11.5 brains (Fig. 3B,E). Next, we carried out a detailed IHC and semiquantification analysis on sagittal sections of E10.5, E11.5, and E12.5 brains. The results were surprising on several accounts. First, we discovered an anterior-to-posterior gradient expression of GPR56 in the basal surface of the neocortex in both E10.5 and E11.5 brains; such a pattern is not apparent within the radial glial cell population (Fig. 4B,E,K). This finding is particularly interesting, as it is the region where preplate neurons reside. The gradient expression pattern, however, dissipated by E12.5 (Fig. 4H). Moreover, there was a parallel gradient pattern in the expression of GPR56 and Tuj1 in E10.5 embryonic brains (Fig. 5A–C). By E11.5, Tuj1-positive neurons covered the entire dorsal surface of the embryonic brain. However, only the rostral one-third of the postmitotic neurons were co-labeled with GPR56 (Fig. 5G–O).
CR cells are the major neurons in the preplate and marginal zone of the developing brain. To investigate whether CR cell populations express GPR56, we carried out a series of double immunostainings. We used calretinin and reelin to reveal the CR cell population. At the basal surface of E11.5 mouse brains, CR cells were present sparsely throughout the dorsal telencephalon, whereas dense GPR56 signals were only detected in the rostral one-third of the neocortex (Fig. 6A,D,G,J,M,P). Most of the CR cells in the rostral region of the brain expressed GPR56, whereas the CR cells in the caudal surface of the neocortex were largely negative for GPR56 (Fig. 6H,I,Q,R). To further demonstrate that CR cells express GPR56, we performed double immunocytochemistry on 1DIV progenitor cells. As shown in Figure 6V, some CR cells indeed express GPR56.
In our previous study, we observed intense GPR56 mRNA expression in the SVZ, a site for IPCs (Piao et al., 2004). Tbr2, a T-domain transcription factor, is highly expressed in IPCs (Englund et al., 2005). To examine whether IPCs themselves express GPR56, we performed double immunostaining of GPR56 and Tbr2 on E12.5 and E14.5 mouse embryonic brains. As expected, Tbr2-positive IPCs strongly express GPR56, suggesting a potentially uncharacterized role for GPR56 in cerebral cortical development (Fig. 7). This observation is further confirmed by double immunocytochemistry on 1DIV progenitor cells (Fig. 7L).
Corticogenesis starts with the first asymmetric division of neural progenitor cells, producing the first wave of postmitotic neurons that migrate along the radial glia processes toward the pia matter. The first-born neurons form the preplate. The second wave splits the preplate into a superficial marginal zone and a deeper subplate, forming the cortical plate in between. Subsequently, the cortical plate expands in an “inside-out” manner, in which newer neurons migrate past the early-born neurons to occupy progressively more superficial layers of the cortex (Ayala et al., 2007; Bystron et al., 2008; Caviness and Rakic, 1978; Gaitanis and Walsh, 2004; Marin and Rubenstein, 2003).
Mutations in GPR56 result in cortical malformation with a decreasing anterior-to-posterior gradient in severity, the mechanism of which was not clear. In this study, we demonstrated that the preplate neurons in the mouse neocortex at E10.5 and E11.5 express GPR56 in an anterior-to-posterior gradient, which is highly correlated with the anatomic distribution of the cortical malformation found in both humans and mice associated with GPR56 mutations. Similarly, our previous work revealed that the expression pattern of GPR56 in cerebellar granule cells accounts for the regional distribution of cerebellar defects associated with Gpr56 gene deletion (Koirala et al., 2009). Taken together, our data suggest that the preplate neurons play an important role in maintaining the integrity of the pial BM and regulating cortical lamination.
IPCs are also called basal progenitors or non-surface-dividing cells (Hevner, 2006). Derived from radial glia, IPCs undergo one or two mitotic cycles to give rise to glutamatergic, pyramidal-projection neurons (Hevner, 2006). IPCs mainly reside in the SVZ of the developing neocortex. Ablation of IPCs by conditional deletion of the mouse Tbr2 in the developing forebrain results in a decrease in cortical surface expansion and thickness as well as neuronal reduction in all cortical layers (Sessa et al., 2008, 2010). Given the fact that GPR56 is implicated in a human cortical malformation and that it shows significant evolutionary differences in its gene structure, it is reasonable to presume that the strong expression of GPR56 in IPCs likely plays a role in regulating the expansion of the cortical surface (Jin et al., 2009; Yu et al., 2009).
In summary, GPR56 is expressed widely in the developing neocortex. How its diverse expression is regulated and what its functions are in different cell lineages, however, remain unknown. In determining the regulation of GPR56 transcription, it is particularly vital to further characterize the promoter region of Gpr56 as well as the contributing transcription factors. By conducting cell type-specific ablation of GPR56, we will be able to study the function of GPR56 in different regions of the developing brain, enabling us to understand how the protein works to regulate brain development as a whole.
Grant sponsor: National Institute of Neurological Disorders and Stroke: Grant number: R01 NS057536 (to X.P.); Grant sponsor: the William Randolph Hearst Fund Award (to S.J.); Grant sponsor: the Leonard and Isabelle Goldenson Research Fellowship (to R.L.).
We thank the Monoclonal Antibody Core, Dana Farber/Harvard Cancer Center (supported in part by National Cancer Institute Cancer Center Support Grant P30 CA06516) for their assistance in generating a monoclonal antibody against GPR56.