SEARCH

SEARCH BY CITATION

Keywords:

  • homeobox gene;
  • somatosensory cortex;
  • monoaminergic neurotransmission;
  • reelin

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Recently, two orthologues of the Drosophila homeobox Cut gene, Cux-1 and Cux-2, have been identified as restricted molecular markers of upper layer (II–IV) neurons in the murine cerebral cortex. We show that during early postnatal life, from P0 to P10, Cux-1 and Cux-2 mRNA are coexpressed in all primary sensory cortices. Antisera to Cux-1 and Cux-2 immunoreactivities preferentially label neurons in the barrel walls of the primary somatosensory cortex (S1). Subsequently, Cux-1 remains enriched in sensory cortices, whereas Cux-2 expression enlarges to comprise the frontal and insular areas. The laminar distribution of Cux-1 and Cux-2 differs: Cux-1 follows a layer IV to layer II decreasing gradient of expression, whereas Cux-2 expression is homogeneous across layers IV–II. No colocalization was found with GABA and birth dating experiments showed that Cux-1–positive neurons in layer IV are born during a restricted period, E13.5–E14.5, suggesting that Cux-1 is a useful molecular marker of the glutamatergic neurons of layer IV. We examined Cux-1 and Cux-2 in barrel-defective mouse strains, the VMAT2 KO, the MAOA KO, and the Adcyl 1brl strain. A normal expression level of Cux-1 and Cux-2 was found in layer IV, despite the lack of segregation of the neurons as barrels. Conversely, in Reeler mice, Cux-1 and Cux-2 had a distinct laminar distribution: the Cux-1–positive neurons had an inverted deep localization, whereas the Cux-2–positive neurons were distributed throughout the cortical thickness, suggesting that Cux-2 expression is more widely expressed in the inverted cortex of reeler mutants. Our results indicate that Cux-1 is a useful marker of the layer IV neurons in S1, and that Cux-1 and Cux-2 are differently regulated in the upper layers of the cerebral cortex. © 2006 Wiley-Liss, Inc.

The cerebral cortex is a six-layered structure comprised primarily of glutamatergic projection neurons, originating from dorsal telencephalic progenitors, and of interneurons of ventral origin. Cortical projection neurons are distinguished early on by their expression of Tbr1 and several bHLH proteins, including NeuroD, Neurogenin1, Neurogenin2, Math2, and Math3 (Fode et al.,2000; Pleasure et al.,2000). It has been proposed that bHLH transcription factors, acting in regulatory cascades, orchestrate neuronal differentiation. However, little is known about the genes that are involved in the specification of laminar identity of principal cortical neurons (Frantz and McConnell,1995).

Recently, two orthologues of the Drosophila homeobox Cut gene, Cux-1 and Cux-2, have been identified as restricted molecular markers of upper layer (II–IV) neurons in the cerebral cortex. Cux-1 and Cux-2 are early markers of neuronal differentiation; both genes are expressed in postmitotic neurons at early embryonic stages. Cux-1 and Cux-2 are also expressed in proliferative zones of the cerebral cortex, with selective Cux-2 expression in SVZ progenitors (Nieto et al.,2004; Zimmer et al.,2004). Interestingly, Pax6 mutant mice show a loss of Cux-2–positive cells in the SVZ as well as a loss of Cux-1 and Cux-2 in the upper layer neurons, supporting the SVZ origin of the glutamatergic neurons in layers II–IV of the neocortex (Nieto et al.,2004; Zimmer et al.,2004).

Cux-1 and Cux-2 are homeobox-containing genes encoding transcription factors. In Drosophila, Cut functions as a determinant of cell type specification in several tissues (Nepveu,2001). Murine Cux and human CDP complement certain Drosophila Cut mutants, suggesting evolutionarily conserved function of Cut proteins (Nepveu,2001). Thus, Cux-1 and Cux-2 are interesting candidate genes for the specification and the maintenance of an upper layer neuronal cell identity.

Here, we describe the dynamic expression of Cux-1 and Cux-2 mRNA and protein in the postnatal somatosensory cortex. We found that Cux-1 has a more restricted expression than Cux-2, which is widely expressed in the frontal areas of the cortex. Furthermore, we show that Cux-1 is expressed in a laminar gradient, whereas Cux-2 is homogeneously distributed throughout the upper layers, suggesting that Cux-1 and Cux-2 have potentially different roles in the differentiation of the upper layer neurons. The level of expression and laminar distribution of Cux-1 and Cux-2 was unchanged in several mutant mice with altered organization of the barrels in the somatosensory cortex. Finally, we found that Cux-2 is ectopically expressed in the Reeler cortex, substantiating the idea of a different regulation of Cux-1 and Cux-2.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Animals

Animal procedures were conducted in strict compliance with approved institutional protocols and in accordance with the provisions for animal care and use described in the European Communities council directive of 24 November 1986 (86/609/EEC). Experiments were carried out on P0, P5, P10, P15, P21, and adult mice. The day of the vaginal plug was counted as E0.5, and the day of birth as P0.

VMAT2 KO mice were generated by a targeted mutation in the first protein codon exon of the VMAT2 gene (Fon et al.,1997). MAOA KO have a deletion of exons 2 and 3 in the gene encoding MAOA (Cases et al., 1996). The barrelless mouse strain (brl) is a spontaneous mutation that was identified as a loss of function of the adenylate cyclase-1 gene (Adcyl 1brl) due to the insertion of an early retrotransposon (Welker et al.,1996; Abdel-Majid et al.,1998). Reeler (Relnorl) mice have a 220 nucleotide deletion in the 3′ region of the Reelin transcript, resulting in a frame shift with the production of a predicted protein amputated of its C-terminal amino acids (de Bergeyck et al.,1997).

Immunocytochemistry

We used antibodies to Cux-1 (1:500; rabbit polyclonal M-222; Santa Cruz, CA) and Cux-2 (1:500; rabbit polyclonal Ab356) (Gringas et al.,2005). The M-222–purified rabbit polyclonal antibody did not give any specific staining on brain sections of Cux-1 mutant mice (not shown). Anesthetized animals were transcardially perfused with saline followed by 4% paraformaldehyde in 0.12 M phosphate buffer, pH 7.4 (PB). Serial sections were cut either on a freezing microtome (free-floating sections; 50 μm) or on a cryostat (15 μm) and immediately processed for immunocytochemistry. In brief, sections were washed in PB and incubated 1 hr in PBS+ (0.1 M PBS with 0.2% gelatin and 0.25% Triton X-100). Sections were incubated sequentially with the primary antibodies (24 hr at 4°C), PBS+ (30 min), secondary antibodies (biotinylated goat antirat; biotinylated swine antirabbit; 1:200; Dako, Denmark; 2 hr at room temperature), PBS+, streptavidin-biotin-peroxidase complex (1:200; Amersham, IL; 2 hr at room temperature), and reacted with a solution containing 0.02% diaminobenzidine, 0.6% nickel ammonium sulfate (Carlo Erba), and 0.003% H2O2 in 0.05 M Tris buffer, pH 7.6. All sections were mounted on TESPA-coated slides (Sigma), air-dried overnight, dehydrated, and coverslipped in DePeX.

Double immunolabeling was performed consecutively with rabbit anti-Cux-1 (1:1,000) and rat anti-GABA (1:500; Affinity Research, U.K.). Free-floating sections were incubated overnight at room temperature with the primary antibodies diluted in PBS+. Then, sections were washed in PB and incubated with the two secondary antibodies (Alexa goat antimouse, 1:200, Molecular Probes, OR; Cy3 goat antirabbit, 1:200, Chemicon, CA).

Birth Dating In Vivo

Pregnant females of the Swiss genetic background received a single BrdU injection (IP; 50 mg/kg; in 0.9 % NaCl) at gestational days E13.5 (n = 2), E14.5 (n = 2), E15.5 (n = 2), and E16.5 (n = 2). The day of the vaginal plug detection was counted as E0.5. Pups were kept until P25 and perfused transcardially with 4% PFA. Brains were removed, postfixed 1 hr in the same fixative, and cryoprotected at least 1 night in freshly made 10% sucrose in PB. Brains were frozen in isopentane and cut coronally on a cryostat (17 μm). Sections were first processed for Cux-1 IR immunofluorescence, as described above, and were then immediately processed to detect BrdU. Sections were treated with 2 N HCl for 45 min, rinsed three times in 0.1 M PBS, pH 7.4, incubated 1 hr in 10% NGS in PBS, and incubated overnight in anti-BrdU (1:100; Progen, Germany) in PBS. After washing in PBS, sections were incubated 2 hr in Alexa 488 goat antimouse antibody. Sections were washed 2 hr in PBS, mounted in Vectashield containing DAPI (Vector), and observed with a fluorescent microscope (Leica, DMR).

To determine the number of BrdU-labeled cells, a minimum of six sections taken through S1 was selected. On each section, layer IV and layers II–III were identified by nuclear staining with DAPI (Vectashield; Vector Laboratories, Burlingame, CA), and the number of Cux-1 and BrdU double-immunolabeled was determined using 40× and 100× objectives.

In Situ Hybridization With Digoxigenin-Labeled Riboprobes

Murine Cux-1 probe corresponds to nt 318–4575 and murine Cux-2 probe to nt 67–4829. The in vitro transcription was carried out using the Promega kit and the probes were labeled with digoxigenin-11-dUTP (Roche Diagnostics). Tissue sections were hybridized with digoxigenin-labeled riboprobes. The frozen sections were postfixed 10 min in 4% PFA, washed in PBS, treated with proteinase K (10 μg/ml; Invitrogen) for 7 min, postfixed 5 min in 4% PFA, washed in PBS, acetylated, washed in PBS 1% Triton X100. Sections were incubated 2 hr at room temperature with hybridization buffer (50% formamide, 5 × SSC, 1 × Denhardt's, 250 μg/ml yeast tRNA, and 500 μg/ml herring sperm DNA), hybridized overnight at 72°C with riboprobes (1:200), rinsed for 2 hr in 2 × SSC at 72°C, and blocked in 0.1 M Tris, pH 7.5, 0.15 M NaCl (B1) containing 10% normal goat serum (NGS) for 1 hr at room temperature. After blocking, tissue sections were incubated overnight at room temperature with anti-DIG-antibody conjugated with the alkaline phosphatase (1:5,000; Roche Diagnostics) in B1 containing 1% NGS. After washing in B1 buffer, the alkaline phosphatase activity was detected with nitroblue-tetrazolium chloride (337.5 μg/ml) and 5-bromo-4-chloro-3-indolyl phosphate (175 μg/ml). Sections were mounted in Mowiol (Calbiochem, Carlstadt, Germany).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Cux-1 and Cux-2 mRNA Expression in Postnatal Cerebral Cortex

From P0, Cux-1 and Cux-2 mRNA were present in the superficial layers of the developing neocortex and displayed a similar organization. Interestingly, from P5, Cux-1 mRNA displayed a more restricted pattern than Cux-2 mRNA. Whereas Cux-2 was expressed homogeneously from layers IV to II of the somatosensory, and motor frontal cortices (Fig. 1B), Cux-1 mRNA was essentially found in the primary somatosensory, visual, and auditory cortices (Fig. 1A,F). Furthermore, Cux-1 was expressed most intensely in layer IV with a decreasing gradient in layers II and III (Fig. 1C), whereas Cux-2 expression was homogeneous throughout the superficial layers (Fig. 1D). Interestingly, Cux-1 mRNA expression delimited cell clusters in the somatosensory cortex that correspond to the emerging barrels (Fig. 1C) (Woolsey and Van der Loos,1970; Erzurumlu and Kind,2001). In addition, Cux-1 and Cux-2 were coexpressed in scattered neurons of the layer VI of the insular cortex (Fig. 1A and B). From P21, Cux-2 mRNA was also expressed in the cingulate cortex (Fig. 1F) (Nieto et al.,2004). A strong expression of Cux-2 was also noted at the level of the claustrum (not shown).

thumbnail image

Figure 1. Expression of the Cux-1 and Cux-2 mRNA in the cerebral cortex during postnatal development. In situ hybridization with Cux-1 (A, C, and E) and Cux-2 (B, D, and F) mRNA probes. A–D: Coronal brain sections of a P5 pup. E and F: Coronal brain sections of a P21 pup. A and C: Cux-1 mRNA is expressed almost exclusively in the superficial layers (II–IV) of the parietal cortex. A lower level of expression is found in the frontal and insular cortices. Arrows indicate the borders between the cortical areas. In the parietal cortex, aggregates of stronger expression are observed in layer IV, in the emerging barrels (arrows in C). B: At P5, Cux-2 mRNA is expressed in the same cortical regions as Cux-1, with stronger and more homogeneous expression throughout the superficial layers (D). Scattered Cux-2–positive cells are noted in the deep layers of the insular cortex. E and F: At P21, the distribution of Cux-1 is similar to that observed at P5 (E), whereas Cux-2 expression expands to the superficial layers within the frontal cortex and cingulate cortices (F). Ins, insular cortex; Par, parietal cortex; Fr, frontal cortex. Scale bars = 400 μm in A, B, G, and H; 200 μm in C and D; 100 μm in E and F.

Download figure to PowerPoint

Cux-1 and Cux-2 Protein Expression

The specificity of the Cux-2 antiserum has already been described (Gingras et al.,2005). In brief, Cux-2 protein is nuclear and is expressed only in neurons (Quaggin et al.,1996). The spatiotemporal patterns of Cux-1 and Cux-2 IR overlapped entirely that of gene expression (Fig. 2A and B). In particular, Cux-1 was intensely expressed in layer IV of primary sensory (somatosensory, visual, and auditory) and motor cortices from P5 (Fig. 2A and F). In the somatosensory cortex, Cux-1 IR clearly delineated the barrels (Fig. 2C). The reconstruction of serial sections of flattened hemicortices showed the total representation of the barrelfield (Fig. 2E). From P5, Cux-2 IR became more widespread and was uniformly expressed in the superficial layers of the sensory, motor, and insular cortices (Fig. 2B and G). From P21, Cux-2 IR was also found in the superficial layers of the cingulate and limbic cortices (not shown). As observed with the distribution of mRNAs, we found that Cux-1 was strongly expressed by layer IV, whereas Cux-2 was expressed homogeneously through the upper layers (Fig. 2C and D). Nieto et al. (2004) showed that Cux-1 IR was not present in parvalbumin-containing neurons. Parvalbumin labels a minor GABAergic population. Here, we extended the study by showing that Cux-1 was not found in cortical GABA-containing neurons (Fig. 3A and B).

thumbnail image

Figure 2. Localization of the Cux-1 and Cux-2 proteins by immunocytochemistry in the somatosensory cortex. Coronal brain sections (A-D, F, and G) of a P10 pup. A and C: In the primary somatosensory cortex (S1), Cux-1 immunocytochemistry delineates the barrels within layer IV. Scattered Cux-1 IR neurons are found in the supragranular layers II–III. B and D: Cux-2 IR is homogeneously expressed in the superficial layers and delineates barrels in layer IV of S1. E: Reconstruction of the barrelfield on superimposed sections of a flattened hemisphere labeled with Cux-1. F: Caudal brain section showing Cux-1 IR in the primary auditory (A1) and visual cortices (V1). G: Cux-2 IR is homogeneously distributed through the superficial layers of the caudal cortex. As, anterior snout; PMBSF, posteromedial barrel subfield. Scale bars = 400 μm in A and B; 100 μm in C and D; 200 μm in E.

Download figure to PowerPoint

thumbnail image

Figure 3. Neurotransmitter phenotype and birth dating of Cux-1 IR neurons in the layer IV of the cortex. Coronal brain sections of P25 pups. A and B: Double immunostaining of Cux-1 (nuclear) and GABA (cytoplasmic) in S1. B: Higher magnification showing that Cux-1 IR neurons do not contain GABA in layer IV. CF: Double immunostaining of Cux-1 (red) and BrdU (green) in the somatosensory cortex after a single administration of BrdU at E13.5 (C), E14.5 (D), E15.5 (E), or E16.5 (F). G: The histogram shows the mean percentage (SEM) of Cux-1 IR neurons that are colabeled for BrdU versus the total number of Cux-1 IR cells in layer IV (black bars) and layers II–III (white bars). This indicates that Cux-1 IR neurons in layer IV are mainly generated over the E13.5–E14.5 period, whereas Cux-1 IR neurons in layer II–III are generated over the E15.5–E16.5. Scale bars = 100 μm in A; 25 μm in B; 50 μm in C–F.

Download figure to PowerPoint

Birth Date of Cux-1 IR Neurons in Somatosensory Cortex

We next examined the birth date of Cux-1 IR neurons in the somatosensory cortex by injecting BrdU daily during the E13.5–E16.5 period, when most layer II–IV neurons are generated (Fairen et al.,1986). Brains were examined at P25. BrdU injection at E13.5 clearly labeled neurons in the lower part of layer IV and scattered cells in the superficial layers (Fig. 3C). In layer IV, BrdU-positive neurons colocalized with Cux-1 IR (Fig. 3C). Scattered BrdU-positive cells were also located in layers II–III but did not contain Cux-1 IR and may correspond to GABAergic neurons generated in the median ganglionic eminence (Fig. 3C). BrdU injections at E14.5 labeled the upper part of the layer IV and layer III (Fig. 3D). At E15.5, BrdU-positive cells were located in layers II–III and at least 30% were Cux-1 IR (Fig. 3E). At E16.5, BrdU-positive cells were mainly located in layer II and some of them were Cux-1 IR (Fig. 3F). The ratio of Cux-1 and BrdU double IR cells versus Cux-1 IR cells in layer IV and layers II–III at each BrdU injection is shown in Figure 3G. The diagram indicates that Cux-1 IR neurons in layer IV are generated over the E13.5–E14.5 period.

Cux-1 and Cux-2 Expression in Barrel-Defective Mice

We analyzed the expression of the Cux-1 and Cux-2 proteins in the somatosensory cortex of different barrel-defective mouse strains. Mice lacking the vesicular monoamine transporter (VMAT2 KO), the barrelless mutants (Adcyl 1brl), or the monoamine oxidase type A (MAOA KO) have an altered organization of the layer IV neurons, which are distributed as a continuous layer, instead of aggregating in barrels (Cases et al.,1996; Welker et al.,1996; Abdel-Majid et al.,1998; Alvarez et al.,2002). The Cux-1 and Cux-2 proteins were normally expressed in the upper cortical layers of VMAT2, barrelless, and MAOA KO mice (Fig. 4) and correlated to the altered barrel organization in these mice. In the layer IV of S1, Cux-1 or Cux-2 IR neurons form a uniform layer instead of forming barrels (Fig. 4C, D, and G–J).

thumbnail image

Figure 4. Cux-1 and Cux-2 proteins in barrelless mutants. Coronal brain sections. In P5 VMAT2 KO pups, Cux-1 (A) and Cux-2 (B) are normally located in the superficial layers of the primary somatosensory cortex (S1). In layer IV, Cux-1 (C) and Cux-2 (D) immunolabeled neurons are uniformly distributed and do not organize into barrels as in the controls (see Fig. 2). In P7 AdcyI KO pup, Cux-1 (E and G) and Cux-2 (F and H) are arranged abnormally in layer IV. Granular neurons do not segregate in barrels. In MAOA KO mice, Cux-1 (I) and Cux-2 (J) display also a barrelless phenotype. Scale bars = 200 μm in A, B, E, and F; 100 μm in C, D, and G–J.

Download figure to PowerPoint

Cux-1 and Cux-2 Expression in Reeler Mouse

We next analyzed the expression of Cux-1 and Cux-2 proteins in the cerebral cortex of the Reeler mice at P21, which display a partially inverted layering of the somatosensory cortex (Caviness and Sidman,1973; Polleux et al.,1998). Reeler mice display a very poor lamination and have defects of neuronal migration: early-born neurons tend to localize in the upper cortical plate, whereas the late-born neurons tend to be located in the deeper cortical laminae. In agreement with previous descriptions by Nieto et al. (2004), we found that most Cux-1 IR neurons were localized in the deeper layers of S1 (Fig. 5A and C). Thus, the distribution of Cux-1 IR is consistent with the abnormal pattern of migration observed in Reeler somatosensory cortex. Surprisingly, however, the distribution of Cux-2 IR was different from the expected patterns. Indeed, the Cux-2 IR neurons were found throughout the cortical thickness of the frontal and somatosensory cortices (Fig. 5B and D). We observed an area of stronger IR in the deeper part of the somatosensory cortex that could correspond to the upper layers (Fig. 5D).

thumbnail image

Figure 5. Expression of Cux-1 and Cux-2 proteins in Reeler cortex. Coronal brain sections of a P21 pup. A and C: Cux-1 immunoreactivity is found in the late-born neurons located in the deeper layers of the Reeler parietal cortex. B and D: In contrast, Cux-2 immunoreactivity is abnormally distributed throughout the cortical layers of the parietal and frontal cortices. Scale bars = 400 μm in A and B; 100 μm in C and D.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

Cux-1 and Cux-2 are two molecular markers of the cortical upper layer neurons and their precursors (Nieto et al.,2004; Zimmer et al.,2004). Previous descriptions (Nieto et al.,2004; Zimmer et al.,2004) have focused on embryonic development, suggesting a potential role of Cux-1 and Cux-2 in the acquisition and maintenance of cell identity of the upper layers of the cerebral cortex. Here, we describe the postnatal expression of Cux-1 and Cux-2 mRNA and protein in the somatosensory cortex of normal and barrelless mutant mice.

Cux-1 as Valuable Marker of Somatosensory Cortex

From P0, Cux-1 and Cux-2 mRNA and protein are strongly expressed in accumulating layer IV neurons. From P5, we observe that Cux-1 mRNA and protein is expressed in a laminar gradient, with layer IV showing the strongest level of expression. In contrast, Cux-2 is uniformly expressed in most of the upper layer neurons. Interestingly, on serial sections of a normal flattened cortex, Cux-1 IR enables the reconstruction of the entire barrelfield, proving Cux-1 IR as a valuable tool to examine the integrity of S1 in genetic models. Furthermore, Cux-1 has a more restricted pattern of expression than Cux-2. Indeed, Cux-1 is specifically expressed in the sensory cortical areas, whereas Cux-2 has a broader expression including the cingulate and limbic areas by P10–P15. This suggests that Cux-1 and Cux-2 could have complementary roles in the differentiation of the cerebral cortex, rather than being redundant.

Upper layer cortical neurons are generated over the E13.5–E16.5 period and the genesis of layer IV neurons is restricted to the E13.5–E14.5 period (Fairen et al.,1986). We show that Cux-1 is expressed in layer IV neurons that are generated between E13.5 and E14.5. Moreover, the Cux-1–positive neurons do not contain GABA, indicating that they correspond to the principal glutamatergic cell type of neurons in layer IV. This complements previous observations that showed Cux-2 in a neuronal population distinct from the parvalbumin-containing neurons, a marker of a subpopulation of GABAergic neurons (Nieto et al.,2004). Our observations thus identify Cux-1 as a marker of the granular neurons of layer IV in S1.

Cux-1 and Cux-2 Expression in Mutants With an Altered Barrelfield

The spatiotemporal expression of Cux-1 and Cux-2 in the somatosensory cortex from P0 suggests a potential role of these transcription factors in acquiring and maintaining granular neurons and thalamocortical axon segregation in the barrels. Thalamocortical axons reach the cortex from E15 and target layer IV by P0. Then thalamocortical axons branch profusely to acquire whisker-related patterns by P3–P4 (Agmon et al., 1995; Rebsam et al.,2002). From P4, granular neurons aggregate around thalamocortical axons to form barrels (Woolsey and Van der Loos,1970; Erzurumlu and Kind,2001). In recent years, several mutant mouse models lacking cortical barrels have began to unveil the cellular and molecular mechanisms by which these patterns emerge. Neural activity plays a crucial role in conferring presynaptic patterns to postsynaptic cells via neurotransmitter receptor-mediated intracellular signals (Erzurumlu and Kind,2001; Laurent et al.,2002). MAOA KO have an excess level of brain serotonin, causing a lack of segregation of both thalamocortical and cortical neurons (Cases et al.,1996). VMAT2 KO have a drastic reduction of monoaminergic neurotransmitters, serotonin, dopamine, and noradrenalin in the brain and display an altered organization of the layer IV neurons, although thalamocortical axons segregate almost normally (Alvarez et al.,2002). Adcyl 1 is expressed all along the primary sensory pathway (Nicol et al.,2005) and barrelless mice Adcyl 1brl display a lack of segregation of both thalamocortical axons and cortical neurons (Welker et al.,1996). In these barrelless mutants, Cux-1 and Cux-2 IR display a normal laminar expression. In layer IV, despite an altered barrelfield organization, Cux-1 and Cux-2 appear to be expressed at a normal level, suggesting that Cux-1 and Cux-2 expression could not be dependent on extrinsic factors such as monoaminergic neurotransmission or cAMP-dependent mechanisms. Further experiments including Western blots are necessary to support this hypothesis. It will be interesting to determine whether Cux-1 and Cux-2 expression is controlled by extrinsic afferents such as the thalamocortical axons.

Cux-1 and Cux-2 Expression in Reeler Cortex

Early reports in Reeler claimed that the early-generated neurons are destined to the superficial cortical layers, whereas late-generated neurons are destined to the deep cortical layers (Caviness and Sidman,1973). In fact, the inversion of the normal histogenetic gradient is found to a limited extent in the somatosensory cortex, whereas a considerable radial intermixing of different cortical populations is characteristic of the frontal areas (Terashima et al.,1983; Inoue et al.,1991; Polleux et al.,1998). Indeed, Cux-1 IR, a marker of layer IV in the adult somatosensory cortex is grossly inverted in Reeler, although a normal layering is not observed. Cux-2 is expressed in the upper layer neurons of most cortical areas, and then we should expect a relatively inverted distribution in the somatosensory cortex and a considerable radial intermixing in the other cortical areas. In fact, we observe a profound and unexpected change of Cux-2 IR. In reeler mice, Cux-2–positive cells are observed throughout the cortical laminas and in all cortical areas. At this point of the study, it is difficult to determine whether the abnormally widespread expression of Cux-2 cells is a consequence of a radial intermixing of the upper layer neurons or a consequence of a modification of the proliferation dynamics of cortical progenitors that has been observed in Reeler. During embryonic development, Cux-1 and Cux-2 expression differs in cortical progenitors. Cux-2 is expressed in an early pool (E11.5) of cells that divide subsequently in the subventricular zone to give rise ultimately to the upper layers (Zimmer et al.,2004), whereas Cux-1 is expressed later in SVZ precursors that subsequently give rise to the upper layers (Nieto et al.,2004). This suggests that Cux-2 may have more importance for the proliferation dynamics of cortical precursors than Cux-1, which might act as a differentiation factor of the upper layer neurons. Then as a consequence, Cux-1 IR appears normally inverted in Reeler somatosensory cortex, whereas Cux-2 IR shows a more complex pattern.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED

The authors thank Alain Nepveu, Xavier Nicol, Alexandra Rebsam, Léa Stankovski, Egbert Welker, and Marion Wassef for providing biological materials, and Chantal Alvarez for technical help.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. LITERATURE CITED