- Email: [email protected]
Expression of hyaluronan (hyaluronic acid) in the developing laminar architecture of the human fetal brain
Annals of Anatomy 195 (2013) 424–430
Contents lists available at ScienceDirect
Annals of Anatomy journal homepage: www.elsevier.de/aanat
Research article
Expression of hyaluronan (hyaluronic acid) in the developing laminar architecture of the human fetal brain Shunichi Shibata a , Kwang Ho Cho b,∗ , Ji Hyun Kim c , Hiroshi Abe d , Gen Murakami e , Baik Hwan Cho f a
Maxillofacial Anatomy, Department of Maxillofacial Biology, Tokyo Medical and Dental University Graduate School, Tokyo, Japan Department of Neurology, Wonkwang University School of Medicine, Jeonbuk Regional Cardiocerebrovascular Disease Center, Institute of Wonkwang Medical Science, 895, Muwang-ro, Iksan, Jeonbuk 570-711, Republic of Korea c Department of Anatomy, Chonbuk National University College of Medicine, Jeonju, Republic of Korea d Department of Anatomy, Akita University School of Medicine, Akita, Japan e Division of Internal Medicine, Iwamizawa Kojin-kai Hospital, Iwamizawa, Japan f Department of Surgery, Chonbuk National University College of Medicine, Jeonju, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 15 April 2013 Received in revised form 9 July 2013 Accepted 15 July 2013
Keywords: Hyaluronan Subplate Cajal–Retzius cells Cortical stratified transitional field Neocortex Human fetus
s u m m a r y Hyaluronan (also called hyaluronic acid or HA) plays a key role in the morphogenesis of the brain, but little is known about its expression in the human fetal neocortex. Using immunohistochemical methods, we assayed the expression of HA, glial fibrillary acidic protein, vimentin, nestin, and proliferating cell nuclear antigen in paraffin-embedded histologic sections of 8 mid-term fetuses (estimated gestational age, 12–16 weeks; crown-rump length, 75–120 mm). At 12–13 weeks, HA was expressed strongly along the membranes of many cells in the cortical plate and the layer 1 or marginal zone, but showed weak, spotty expression in a fiber-rich layer adjacent to the cortical plate, called the cortical stratified transitional field-1 (STF-1 or a primitive form of the subplate). At 15–16 weeks, HA was expressed in the layer 1 and in the early subplate or presubplate, but less strongly in cells of the possible STF-5 near the subventricular zone. However, the positive observation in STF-5 was probably a result of individual difference in development. The developing cortical plate seemed to produce HA in the presubplate to harbor axonal plexus of various afferent systems, while Cajal–Retzius cells were likely to accumulate HA in the layer 1. The HA-rich zones, those sandwiched the cortical plate, might avoid further migration of cortical cells. © 2013 Elsevier GmbH. All rights reserved.
1. Introduction Hyaluronan (also called hyaluronic acid or HA), a nonsulfated linear glycosaminoglycan, has been found to facilitate cell movement in the fetal brain by weakening cell attachment to adhesive substrates and by creating hydrated pathways for migrating cells (reviewed by Bignami et al., 1993). To our knowledge, however, HA distribution has not been assessed in the developing laminar or stratified architecture of the fetal cerebrum in any mammal (Meyer et al., 2000; Bayatti et al., 2008; Clowry et al., 2010). Limited information is available on the laminar distribution of proteoglycans in the visual cortex of adult cats (granular cell layer; Lander et al., 1997) and in the telencephalic vesicle of a 17-day rat fetus (Bignami et al., 1993).
∗ Corresponding author at: Department of Neurology, Wonkwang University School of Medicine and Hospital, 895, Muwang-ro, Iksan, Jeonbuk, 570-711, Republic of Korea. Tel.: +82 63 859 1411; fax: +82 63 842 7379. E-mail addresses: [email protected], [email protected] (K.H. Cho). 0940-9602/$ – see front matter © 2013 Elsevier GmbH. All rights reserved. http://dx.doi.org/10.1016/j.aanat.2013.07.002
Recently, in the postnatal development and adult morphology of the brain and spinal cord, many research groups have paid attention to HA as a major component of perineuronal nets that isolate synapses and hinder lateral diffusion of postsynaptic receptors to control or stabilize the synaptic plasticity (Frischknecht and Seidenbecher, 2008; Kwok et al., 2011). HA in perineuronal nets is anchored to HA synthase-3 on the cell membrane of interneurons (Giamanco and Matthews, 2012). During postnatal development, however, other fiber-like HA structures distinct from perineuronal nets were observed in the putative white matter of mouse cerebellum (Baier et al., 2007). These findings suggested that HA-containing extracellular structures were not limited to perineuronal nets, especially during fetal brain development. We therefore investigated whether HA structures, other than perineuronal nets, are present in the human fetal neocortex. We also analyzed the topographical relationship between HA expression and cortical laminar structure, by staining tissue samples with antibodies to glial fibrillary acidic protein (GFAP), vimentin, nestin and proliferating cell nuclear antigen (PCNA). According to excellent atlases by Bayer and Altman (2005), in the fetal human neocortex at 12–16 weeks of gestation, the laminar
S. Shibata et al. / Annals of Anatomy 195 (2013) 424–430
structure or cortical stratified transitional field (STF) are identified as follows: (1) the layer 1: the most superficial layer facing the primitive meningis; (2) the cortical plate; (3) STF 1: the superficial fibrous layer or the putative subcortical white matter (possibly corresponding to the primitive form of the subplate in Ulfig et al. (2000), Bystron et al. (2008), and Judaˇs et al. (2010); (4) STF 2: the upper cellular layer or the last sojourn zone before cells translocate to the cortical plate; (5) STF 3: the honeycomb trilaminar matrix of cells and fibers only in granular cortices; (6) STF 4: the complex middle layer or the putative deep white matter; (7) STF 5: the deep cellular layer or the first sojourn zone to appear outside the germinal matrix; (8) STF 6: the late-forming deep layer of callosal fibers outside the germinal matrix; (9) the subventricular zone or SVZ; (10) neuroepithelium or NEP.
425
HA-binding proteins appear in the late stage fetus, near the time of myelination (Bignami et al., 1993). Therefore, HA staining was performed using a biotinylated HA-binding protein (2 g/ml; Seikagaku Corp., Tokyo, Japan) after immersing the sections in chondroitinase ABC (10 microunits/ml; Sikagaku Corp, Tokyo, Japan) in 0.1 M Tris–acetate buffer (pH 8.0, 37 ◦ C) for 30 min (Shibata et al., 2003). The sections were incubated for 30 min with the Histofine SAB kit (Nichirei) for the 3-amino-9-ethylcarbazole (AEC) reaction, yielding a red biotin complex or with Histofine Simple Stain Max-PO (Nichirei) for the DAB reaction with HRP, yielding a dark brown biotin complex. The latter sections were counterstained with hematoxylin.
3. Results 2. Materials and methods Paraffin-embedded specimens were utilized, obtained from 8 mid-term fetuses of estimated gestational age 12–16 weeks and crown-rump length (CRL) of 75–120 mm, including 3 fetuses of gestational age 12–13 weeks and 5 of gestational age 15–16 weeks. These fetuses, obtained by induced abortion, had been donated by the mothers and their families to the Department of Anatomy of Chonbuk National University in Korea, after the mothers were personally informed by an obstetrician about the possibility of donating the fetus for research; no attempt was made to encourage donation. Because of randomization of specimen numbering, it was not possible to trace any of the families concerned. Use of these fetal specimens for research was approved by the ethics committee of Chonbuk National University, which did not require that the corresponding committee in Japan be informed about this research project. This study was performed in accordance with the provisions of the Declaration of Helsinki 1995 (as revised in Edinburgh 2000). Each donated fetus was fixed in 10% w/w neutral formalin solution for more than 1 month; and divided into the head and neck, the thorax, the abdomen, the pelvis and the four extremities. All of these body parts were decalcified by incubating them at 4 ◦ C in 0.5mol/l EDTA (pH 7.5; decalcifying solution B; Wako, Tokyo) for 1–3 days, depending on the size of the specimen. The head and neck specimens were sectioned sagittally or horizontally at 20–50 m intervals, depending on the size of the sections. Sections included not only the brain but also the surrounding structures, including eyes and ears. Thus, each sample included skull base cartilage, a positive control for HA staining. Most sections were stained with hematoxylin and eosin (HE), with others used for immunohistochemistry. The primary antibodies used for immunohistochemical staining were (1) rabbit polyclonal anti-human GFAP (1:100; Dako Cytomation, Kyoto, Japan; catalog number Z0334); (2) mouse monoclonal anti-human vimentin (1:10; Dako, Glostrup, Denmark; catalog number M7020); (3) mouse monoclonal anti-human nestin (1:100; Santa Cruz Biotechnology, Santa Cruz, CA, US; catalog number sc23927); (4) mouse monoclonal anti-human PCNA (1:1000; Abcam, Cambridge, UK); (5) mouse monoclonal anti-human growth associated protein-43 or GAP 43 (1:8000; Sigma-Aldrich, St. Louis, US); and (6) rabbit polyclonal anti-human calretinin (1:100; Invitrogen, CA, US). Samples were not pretreated in the autoclave because of the loose nature of the fetal tissues. Following incubation with primary antibody, the sections were incubated with horseradish peroxidase (HRP)-labeled secondary antibody (Histofine Simple Stain Max-PO, Nichirei, Tokyo) for 30 min, followed by incubation with diaminobenzidine (Histofine Simple Stain DAB, Nichirei) for 3–5 min. All samples were counterstained with hematoxylin.
Fibers positive for vimentin and nestin were observed running across the neocortex, from the deep neuroepithelium to the superficial amorphous layer (layer 1), in both smaller (12–13 weeks; Fig. 1) and larger (15–16 weeks; Figs. 2 and 3) fetuses. Vimentinpositive fibers, indicative of radial glial cells, were clearly observed (Figs. 1C, and 2C and G) because the fibers reached the cortical plate. However, the fibers did not express GFAP in the cortical plate of smaller fetuses (Fig. 1B). We also attempted to identify stratified structures in the neocortex according to Bayer and Altman (2005). Layer 1, the cortical plate, the subventricular zone and the neuroepithelium were evident in both larger and smaller fetuses. Layer 1 showed strong expression of HA (Fig. 1A and 2F). In the smaller fetuses (Fig. 1), strong HA expression was observed along the membranes of many cells in the cortical plate and the early subplate, especially in the latter, with weak spotty expression observed in a fiber-rich intermediate layer adjacent to the cortical plate. Likewise, in layer 1, HA expression was not diffuse but appeared to be concentrated around composite cells. Thus, diffuse expression of HA in layer 1 and the subplate appeared to occur around 14 weeks. These HA-binding cells in the cortical plates of smaller fetuses did not express either nestin or GFAP. HA-positive cells, apparent in the putative STF 5 of the larger fetus (see below), were not observed in smaller fetuses. Thus, the other cortical stratified transitional fields (STFs 2–6) could not be distinguished in the smaller fetuses. The early subplate contained axon bundles which were strongly positive for GAP43 (Fig. 1F). Calretinin-positive cells were almost evenly distributed throughout all layers of the neocortex (Fig. 3B). In the larger fetuses, the subplate or STF 1, the superficial fibrous layer or putative subcortical white matter, was characterized by (1) diffuse expression of HA (Fig. 2F), (2) strong expression of GAP 43 (Fig. 2B), (3) few GFAP-positive fibers (Fig. 2G) and (4) absence of PCNA-positive cells (Fig. 2D). If, in larger fetuses, STF 5, consisting of the deep cellular layer or the first transitional zone to appear outside the germinal matrix, was defined as a thin layer containing HA-bound cells in AEC reaction (Fig. 2F), then STF 2–4 and 6 could be identified based on the different arrangements of GFAP-positive fibers between layers (Fig. 2G). Nevertheless, DAB reaction did not demonstrate HA-positive cells in the probable STF 5 despite the strong positivity in the layer 1 and subplate (Fig. 3A). Moreover, even with AEC reaction, the positive cells in the STF5 were not found in our later trial 6 months after the sectioning (figures, not shown). Calretinin-positive cells were rarely seen and, if present, mostly distributed in the STF 2–6 or the intermediate zone (Fig. 3C). Laminar expression of HA was not observed in the large olfactory bulb, the thalamus, the ganglionic eminence (the putative basal ganglia) or the brainstem of both larger and smaller fetuses (Fig. 4). In sites other than the neocortex, HA-positive cells were distributed widely and evenly; we were unable to identify a specific topographical relationship between the putative nucleus and HA expression.
426
S. Shibata et al. / Annals of Anatomy 195 (2013) 424–430
Fig. 1. Expression of hyaluronan and intermediate filaments in the neocortex (putative temporal lobe) of a 12-week human fetus. Nearly horizontal sections. Panels A–F display proximate sections. Hyaluronan (panel A) is colored brown with diaminobenzidine (DAB) reaction. Panel B, glial fibrillary acidic protein (GFAP); panel C, vimentin; panel D, nestin; panel E, proliferating cell nuclear antigen (PCNA); panel F, growth associated protein-43 (GAP 43). All panels are presented at the same magnification (scale bar in panel C). Hyaluronan expression is seen in the layer 1 (or the marginal zone), the cortical plate and the early subplate (subplate). NEP, neuroepithelium; SVZ, subventricular zone.
The developing pyramidal tract was negative for HA (not shown). However, at the stages examined, the hippocampus was not well differentiated, but appeared as a simple fold of the cortex. The morphology of the cerebellum was also primitive, without laminar architecture or folds. 4. Discussion Using an anatomic atlas (Bayer and Altman, 2005), we attempted to identify cortical STFs in mid-term fetuses. The cortical plate was easily identified by its high cell density, whereas the subplate was characterized by the specific results of immunostaining for HA, GAP43, and PCNA. Our identified subplate seemed to correspond to that in Bayatti et al. (2008) and Ip et al. (2011) because of the strong expression of GAP43. However, the differentiation was still in a primitive stage because of few calretinin-positive neurons (present study) as well as because of no or few
nNOS-positive neurons (unpublished data), both of them are dominant populations of the definite subplate (Downen et al., 1999; Ulfig et al., 2000) at 12–16 weeks. Bayatti et al. (2008) also demonstrated a strong expression of synaptophysin in the subplate, but we failed to stain it in the present materials (unpublished data). Although the present subplate corresponded to STF 1, it was difficult to clearly discriminate between STFs 2–6, especially in small fetuses at 12–13 weeks. These layers seemed to correspond to the intermediate zone in Ulfig et al. (2000), Bystron et al. (2008) and Judaˇs et al. (2010). However, strangely, in the ventricular side of the subplate or STF 1, PCNA-positive cells were distributed evenly in the intermediate zone as well as in the ventricular and subventricular zone. In large fetuses at 15–16 weeks, arrangement of GFAP-positive fibers as well as laminar expression of HA suggested possible demarcations between these STFs: our hypothetically identified STFs are shown in Fig. 2G. The subplate is known to express fibronectin, chondroitin-sulfate proteoglycans,
S. Shibata et al. / Annals of Anatomy 195 (2013) 424–430
427
Fig. 2. Expression of hyaluronan and intermediate filaments in the neocortex (putative temporal lobe) of a 16-week human fetus. Sagittal sections. Panels A–E exhibit proximate sections. Hyaluronan (panels A and F) is stained red with the 3-amino-9-ethylcarbazole (AEC) reaction. Panel B, growth associated protein-43 (GAP 43); panel C, vimentin; panel D, proliferating cell nuclear antigen (PCNA); panel E, nestin; panel G, glial fibrillary acidic protein (GFAP). Panel F displays a higher magnification view of panel A. Panels A–E (or panels F and G) are presented at the same magnification (scale bar in panel A or G). Hyaluronan expression is seen in the layer 1 (or the marginal zone), the early subplate (subplate) and some deep cells near the subventricular zone (SVZ). NEP, neuroepithelium.
neural cell adhesion molecule (NCAM) and laminin (reviewed by Ulfig et al., 2000). To our knowledge, this study is the first to show laminar or stratified expression of HA in the human fetal neocortex. Notably, laminar expression was restricted to the neocortex, being absent from other parts of the brain in fetuses of 12–16 weeks. Thus, HA was likely to be committed to form an inside-out-gradient in the neocortex. Despite numerous studies on perineuronal nets, there have been few reports of laminar expression of proteoglycans and none of HA. HA was first expressed in superficially migrating cells,
followed by expression in the cortical plate at 12–13 weeks and later in the most superficial layer (layer 1) as well as in putative subcortical white matter (STF 1 or the subplate). The HA accumulation seemed to occur in accordance with the cortical plate formation. This result was consistent with findings showing that HA-binding glycoprotein was expressed in the cerebral cortex, specifically in the intermediate layer, of fetal rats at 15–19 days of gestation, but was not expressed in the cortical plate (Delpech and Delpech, 1984). The cortical plate cells seemed to produce HA in the STF 1 during their superficial migration to harbor axonal plexus of various
428
S. Shibata et al. / Annals of Anatomy 195 (2013) 424–430
Fig. 3. Expression of hyaluronan after diaminobenzidine (DAB) reaction and distribution of calretinin-positive cells. Sagittal sections. Panel A (16 weeks) displays hyaluronan expression in the neocortex after diaminobenzidine (DAB) reaction. No positive cells are seen in the intermediate zone between the subplate and the subventricular zone (SVZ). Panel B (12 weeks) and panel C (16 weeks) exhibit distribution of calretinin-positive cells. Calretinin-positive cells are few in number in the larger specimen.
afferent systems, while Cajal-Retzius cells were likely to accumulate HA in the layer 1. In addition to the tangential migration, HA-positive cells in the intermediate zone (STF-5 in the present figure) were also likely to be cooperated into the subplate. This possibility of axial migration seems to be consistent with Wang et al. (2010). The HA-rich zones, which sandwiched the cortical plate, might prevent further migration of cortical cells as HA-associated adhesive cues in the hippocampus (Förster et al., 2001). The rich HA in the subplate seemed to induce cell proliferation and differentiation as the other proteoglycans do (Li et al., 2005; Girós et al., 2007). In contrast to the fibrous or amorphous structures of HA in layer 1 and STF 1, spotty expression of HA in the deep layer surrounding or bound to some cells enabled us to identify the deep layer as STF 5. These HA-positive deep cells were interesting, but we did not find them after DAB reaction. Moreover, even with 3-amino-9-ethylcarbazole (AEC) reaction, we failed to stain them
on the old sections 6 months after preparation. Thus, they were likely to be a result of individual difference in development or, simply, an artifact in AEC reaction for an unknown reason. Perineuronal nets are composed of extracellular matrix molecules, including chondroitin sulfate proteoglycans, HA, tenascin-R, and link proteins, which are interposed between glial processes and nerve-cell surfaces in adult brains (Carulli et al., 2006). We did not examine the expression of chondroitin sulfate proteoglycans, but laminar expression of HA was most likely different from HA expression in perineuronal nets. Fiber-like HA structures in the putative white matter of mouse cerebellum were found to develop postnatally, with these structures being in the vicinity of inhibitory interneuron precursors and Purkinje cells (Baier et al., 2007). Although these structures may be regarded as distinct from perineuronal nets, the pattern of HA expression in the cerebellum does not indicate a laminar architecture but a
S. Shibata et al. / Annals of Anatomy 195 (2013) 424–430
429
Fig. 4. Expression of hyaluronan and intermediate filaments in the brain stem of a 12-week human fetus. Nearly horizontal sections. Panels A–C display neighboring sections. Hyaluronan (panel A) is colored brown with diaminobenzidine (DAB) reaction. Panel B, glial fibrillary acidic protein (GFAP); panel C, nestin. All panels are presented at the same magnification (scale bar in panel C). Hyaluronan-positive cells are distributed widely in the section. Nestin-positive fibers are evident from the interpeduncular area to the putative septal nucleus complex (panel C). III, putative oculomotor nucleus complex.
meshwork evenly distributed throughout the putative white matter. Thus, these HA structures seemed different from HA expression in STF 1. Actually, we found that the developing pyramidal tract was negative for HA. Laminar expression of HA in the superficial and deep sides of the cortical plate may result from the differentiation of cortical neurons, inducing a layer of putative white matter. Acknowledgment This study was supported by a grant (0620220-1) from the National R & D Program for Cancer Control, Ministry of Health & Welfare, Republic of Korea.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.aanat. 2013.07.002.
References Baier, C., Baader, S.L., Jankowski, J., Gieselmann, V., Schilling, K., Rauch, U., Kappler, J., 2007. Hyaluronan is organized into fiber-like structures along migratory pathways in the developing mouse cerebellum. Matrix Biol. 26, 348–358. Bayatti, N., Moss, J.A., Sun, L., Ambrose, P., Ward, J.F., Lindsay, S., Clowry, G.J., 2008. A molecular neuroanatomical study of the developing human neocortex from 8
430
S. Shibata et al. / Annals of Anatomy 195 (2013) 424–430
to 17 postconceptional weeks revealing the early differentiation of the subplate and subventricular zone. Cereb. Cortex 18, 1536–1548. Bayer, A.H., Altman, J., 2005. The human brain during the second trimester. In: Bayer, A.H., Altman, J. (Eds.), Atlas of Human Central Nervous System Development Series, vol. 3. CRC Press of Taylor & Francis group, Boca Raton, France, p. 34, 65, 78, 79 (temporal lobe); pp. 100–109 (brain stem). Bignami, A., Hosley, M., Dahl, D., 1993. Hyaluronic acid and hyaluronic acid-binding proteins in brain extracellular matrix. Anat. Embryol., 419–433. Bystron, I., Blakemore, C., Rakic, P., 2008. Development of the human cerebral cortex: Boulder Committee revisited. Nat. Rev. Neurosci. 9, 110–122. Carulli, D., Rhodes, K.E., Brown, D.J., Bonnert, T.P., Pollack, S.J., Oliver, K., Strata, P., Fawcett, J.W., 2006. Composition of perineuronal nets in the adult rat cerebellum and the cellular origin of their components. J. Comp. Neurol. 494, 559–577. Clowry, G., Molnár, Z., Rakic, P., 2010. Renewed focus on the developing human neocortex. J. Anat. 217, 276–288. Delpech, A., Delpech, B., 1984. Expression of hyaluronic acid-binding glycoprotein, hyaluronectin, in the developing rat embryos. Dev. Biol. 101, 391–400. Downen, M., Zhao, M.L., Lee, P., Weidenheim, K.M., Dickson, D.W., Lee, S.C., 1999. Neuronal nitric oxide synthase expression in developing and adult human CNS. J. Neuropathol. Exp. Neurol. 58, 12–21. Förster, E., Zhao, S., Frotscher, M., 2001. Hyaluronan-associated adhesive cues control fiber segregation in the hippocampus. Development 128, 3029–3039. Frischknecht, R., Seidenbecher, C.I., 2008. The crosstalk of hyaluronan-based extracellular matrix and synapses. Neuron Glia Biol. 4, 249–257. Giamanco, K.A., Matthews, R.T., 2012. Deconstructing the perineuronal net: cellular contributions and molecular composition of the neuronal extracellular matrix. Neuroscience 218, 367–384. Girós, A., Morante, J., Gil-Sanz, C., Fairén, A., Costell, M., 2007. Perlecan controls neurogenesis in the developing telencephalon. BMC Dev. Biol. 7, 29.
Ip, B.K., Bayatti, N., Howard, N.J., Lindsay, S., Clowry, G.J., 2011. The corticofugal neuron-associated genes ROBO1, SRGAP1 and CTIP2 exhibit an anterior to posterior gradient of expression in early fetal human neocortex development. Cereb. Cortex 21, 1395–1407. ˙ N., 2010. Populations of subJudaˇs, M., Sedmak, G., Pletikos, M., Jovanov-Miloˇsevic, plate and interstitial neurons in fetal and adult human telencephalon. J. Anat. 217, 381–399. Kwok, J.C., Dick, G., Wang, D., Fawcett, J.W., 2011. Extracellular matrix and perineuronal nets in CNS repair. Dev. Neurobiol. 71, 1073–1089. Lander, C., Kind, P., Maleski, M., Hockfield, S., 1997. A family of activity-dependent neuronal cell-surface chondroitin sulfate proteoglycans in cat visual cortex. J. Neurosci. 17, 1928–1939. Li, H.P., Oohira, A., Ogawa, M., Kawanura, K., Kawano, H., 2005. Aberrant trajectory of thalamocortical axons associated with abnormal localization of neurocan immunoreactivity in the cerebral neocortex of reeler mutant mice. Eur. J. Neurosci. 22, 2689–2696. Meyer, G., Schaaps, J.P., Moreau, L., Goffinet, A.M., 2000. Embryonic and early fetal development of the human neocortex. J. Neurosci. 20, 1858–1868. Shibata, S., Fukada, K., Imai, H., Abe, T., Yamashita, Y., 2003. In situ hybridization and immunohistochemistry of versican, aggrecan, and link protein and histochemistry of hyaluronan in the developing mouse limb bud cartilage. J. Anat. 203, 425–432. Ulfig, N., Neudörfer, F., Bohl, J., 2000. Transient structures of the human fetal brain: subplate, thalamic reticular complex, ganglionic eminence. Histol. Histopathol. 15, 771–790. Wang, W.Z., Hoerder-Suabedissen, A., Oeschger, F.M., Bayatti, N., Ip, B.K., Lindsay, S., Supramaniam, V., Srinivasan, L., Rutherford, M., Møllgård, K., Clowry, G.J., Molnár, Z., 2010. Subplate in the developing cortex of mouse and human. J. Anat. 217, 368–380.
Contents lists available at ScienceDirect
Annals of Anatomy journal homepage: www.elsevier.de/aanat
Research article
Expression of hyaluronan (hyaluronic acid) in the developing laminar architecture of the human fetal brain Shunichi Shibata a , Kwang Ho Cho b,∗ , Ji Hyun Kim c , Hiroshi Abe d , Gen Murakami e , Baik Hwan Cho f a
Maxillofacial Anatomy, Department of Maxillofacial Biology, Tokyo Medical and Dental University Graduate School, Tokyo, Japan Department of Neurology, Wonkwang University School of Medicine, Jeonbuk Regional Cardiocerebrovascular Disease Center, Institute of Wonkwang Medical Science, 895, Muwang-ro, Iksan, Jeonbuk 570-711, Republic of Korea c Department of Anatomy, Chonbuk National University College of Medicine, Jeonju, Republic of Korea d Department of Anatomy, Akita University School of Medicine, Akita, Japan e Division of Internal Medicine, Iwamizawa Kojin-kai Hospital, Iwamizawa, Japan f Department of Surgery, Chonbuk National University College of Medicine, Jeonju, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 15 April 2013 Received in revised form 9 July 2013 Accepted 15 July 2013
Keywords: Hyaluronan Subplate Cajal–Retzius cells Cortical stratified transitional field Neocortex Human fetus
s u m m a r y Hyaluronan (also called hyaluronic acid or HA) plays a key role in the morphogenesis of the brain, but little is known about its expression in the human fetal neocortex. Using immunohistochemical methods, we assayed the expression of HA, glial fibrillary acidic protein, vimentin, nestin, and proliferating cell nuclear antigen in paraffin-embedded histologic sections of 8 mid-term fetuses (estimated gestational age, 12–16 weeks; crown-rump length, 75–120 mm). At 12–13 weeks, HA was expressed strongly along the membranes of many cells in the cortical plate and the layer 1 or marginal zone, but showed weak, spotty expression in a fiber-rich layer adjacent to the cortical plate, called the cortical stratified transitional field-1 (STF-1 or a primitive form of the subplate). At 15–16 weeks, HA was expressed in the layer 1 and in the early subplate or presubplate, but less strongly in cells of the possible STF-5 near the subventricular zone. However, the positive observation in STF-5 was probably a result of individual difference in development. The developing cortical plate seemed to produce HA in the presubplate to harbor axonal plexus of various afferent systems, while Cajal–Retzius cells were likely to accumulate HA in the layer 1. The HA-rich zones, those sandwiched the cortical plate, might avoid further migration of cortical cells. © 2013 Elsevier GmbH. All rights reserved.
1. Introduction Hyaluronan (also called hyaluronic acid or HA), a nonsulfated linear glycosaminoglycan, has been found to facilitate cell movement in the fetal brain by weakening cell attachment to adhesive substrates and by creating hydrated pathways for migrating cells (reviewed by Bignami et al., 1993). To our knowledge, however, HA distribution has not been assessed in the developing laminar or stratified architecture of the fetal cerebrum in any mammal (Meyer et al., 2000; Bayatti et al., 2008; Clowry et al., 2010). Limited information is available on the laminar distribution of proteoglycans in the visual cortex of adult cats (granular cell layer; Lander et al., 1997) and in the telencephalic vesicle of a 17-day rat fetus (Bignami et al., 1993).
∗ Corresponding author at: Department of Neurology, Wonkwang University School of Medicine and Hospital, 895, Muwang-ro, Iksan, Jeonbuk, 570-711, Republic of Korea. Tel.: +82 63 859 1411; fax: +82 63 842 7379. E-mail addresses: [email protected], [email protected] (K.H. Cho). 0940-9602/$ – see front matter © 2013 Elsevier GmbH. All rights reserved. http://dx.doi.org/10.1016/j.aanat.2013.07.002
Recently, in the postnatal development and adult morphology of the brain and spinal cord, many research groups have paid attention to HA as a major component of perineuronal nets that isolate synapses and hinder lateral diffusion of postsynaptic receptors to control or stabilize the synaptic plasticity (Frischknecht and Seidenbecher, 2008; Kwok et al., 2011). HA in perineuronal nets is anchored to HA synthase-3 on the cell membrane of interneurons (Giamanco and Matthews, 2012). During postnatal development, however, other fiber-like HA structures distinct from perineuronal nets were observed in the putative white matter of mouse cerebellum (Baier et al., 2007). These findings suggested that HA-containing extracellular structures were not limited to perineuronal nets, especially during fetal brain development. We therefore investigated whether HA structures, other than perineuronal nets, are present in the human fetal neocortex. We also analyzed the topographical relationship between HA expression and cortical laminar structure, by staining tissue samples with antibodies to glial fibrillary acidic protein (GFAP), vimentin, nestin and proliferating cell nuclear antigen (PCNA). According to excellent atlases by Bayer and Altman (2005), in the fetal human neocortex at 12–16 weeks of gestation, the laminar
S. Shibata et al. / Annals of Anatomy 195 (2013) 424–430
structure or cortical stratified transitional field (STF) are identified as follows: (1) the layer 1: the most superficial layer facing the primitive meningis; (2) the cortical plate; (3) STF 1: the superficial fibrous layer or the putative subcortical white matter (possibly corresponding to the primitive form of the subplate in Ulfig et al. (2000), Bystron et al. (2008), and Judaˇs et al. (2010); (4) STF 2: the upper cellular layer or the last sojourn zone before cells translocate to the cortical plate; (5) STF 3: the honeycomb trilaminar matrix of cells and fibers only in granular cortices; (6) STF 4: the complex middle layer or the putative deep white matter; (7) STF 5: the deep cellular layer or the first sojourn zone to appear outside the germinal matrix; (8) STF 6: the late-forming deep layer of callosal fibers outside the germinal matrix; (9) the subventricular zone or SVZ; (10) neuroepithelium or NEP.
425
HA-binding proteins appear in the late stage fetus, near the time of myelination (Bignami et al., 1993). Therefore, HA staining was performed using a biotinylated HA-binding protein (2 g/ml; Seikagaku Corp., Tokyo, Japan) after immersing the sections in chondroitinase ABC (10 microunits/ml; Sikagaku Corp, Tokyo, Japan) in 0.1 M Tris–acetate buffer (pH 8.0, 37 ◦ C) for 30 min (Shibata et al., 2003). The sections were incubated for 30 min with the Histofine SAB kit (Nichirei) for the 3-amino-9-ethylcarbazole (AEC) reaction, yielding a red biotin complex or with Histofine Simple Stain Max-PO (Nichirei) for the DAB reaction with HRP, yielding a dark brown biotin complex. The latter sections were counterstained with hematoxylin.
3. Results 2. Materials and methods Paraffin-embedded specimens were utilized, obtained from 8 mid-term fetuses of estimated gestational age 12–16 weeks and crown-rump length (CRL) of 75–120 mm, including 3 fetuses of gestational age 12–13 weeks and 5 of gestational age 15–16 weeks. These fetuses, obtained by induced abortion, had been donated by the mothers and their families to the Department of Anatomy of Chonbuk National University in Korea, after the mothers were personally informed by an obstetrician about the possibility of donating the fetus for research; no attempt was made to encourage donation. Because of randomization of specimen numbering, it was not possible to trace any of the families concerned. Use of these fetal specimens for research was approved by the ethics committee of Chonbuk National University, which did not require that the corresponding committee in Japan be informed about this research project. This study was performed in accordance with the provisions of the Declaration of Helsinki 1995 (as revised in Edinburgh 2000). Each donated fetus was fixed in 10% w/w neutral formalin solution for more than 1 month; and divided into the head and neck, the thorax, the abdomen, the pelvis and the four extremities. All of these body parts were decalcified by incubating them at 4 ◦ C in 0.5mol/l EDTA (pH 7.5; decalcifying solution B; Wako, Tokyo) for 1–3 days, depending on the size of the specimen. The head and neck specimens were sectioned sagittally or horizontally at 20–50 m intervals, depending on the size of the sections. Sections included not only the brain but also the surrounding structures, including eyes and ears. Thus, each sample included skull base cartilage, a positive control for HA staining. Most sections were stained with hematoxylin and eosin (HE), with others used for immunohistochemistry. The primary antibodies used for immunohistochemical staining were (1) rabbit polyclonal anti-human GFAP (1:100; Dako Cytomation, Kyoto, Japan; catalog number Z0334); (2) mouse monoclonal anti-human vimentin (1:10; Dako, Glostrup, Denmark; catalog number M7020); (3) mouse monoclonal anti-human nestin (1:100; Santa Cruz Biotechnology, Santa Cruz, CA, US; catalog number sc23927); (4) mouse monoclonal anti-human PCNA (1:1000; Abcam, Cambridge, UK); (5) mouse monoclonal anti-human growth associated protein-43 or GAP 43 (1:8000; Sigma-Aldrich, St. Louis, US); and (6) rabbit polyclonal anti-human calretinin (1:100; Invitrogen, CA, US). Samples were not pretreated in the autoclave because of the loose nature of the fetal tissues. Following incubation with primary antibody, the sections were incubated with horseradish peroxidase (HRP)-labeled secondary antibody (Histofine Simple Stain Max-PO, Nichirei, Tokyo) for 30 min, followed by incubation with diaminobenzidine (Histofine Simple Stain DAB, Nichirei) for 3–5 min. All samples were counterstained with hematoxylin.
Fibers positive for vimentin and nestin were observed running across the neocortex, from the deep neuroepithelium to the superficial amorphous layer (layer 1), in both smaller (12–13 weeks; Fig. 1) and larger (15–16 weeks; Figs. 2 and 3) fetuses. Vimentinpositive fibers, indicative of radial glial cells, were clearly observed (Figs. 1C, and 2C and G) because the fibers reached the cortical plate. However, the fibers did not express GFAP in the cortical plate of smaller fetuses (Fig. 1B). We also attempted to identify stratified structures in the neocortex according to Bayer and Altman (2005). Layer 1, the cortical plate, the subventricular zone and the neuroepithelium were evident in both larger and smaller fetuses. Layer 1 showed strong expression of HA (Fig. 1A and 2F). In the smaller fetuses (Fig. 1), strong HA expression was observed along the membranes of many cells in the cortical plate and the early subplate, especially in the latter, with weak spotty expression observed in a fiber-rich intermediate layer adjacent to the cortical plate. Likewise, in layer 1, HA expression was not diffuse but appeared to be concentrated around composite cells. Thus, diffuse expression of HA in layer 1 and the subplate appeared to occur around 14 weeks. These HA-binding cells in the cortical plates of smaller fetuses did not express either nestin or GFAP. HA-positive cells, apparent in the putative STF 5 of the larger fetus (see below), were not observed in smaller fetuses. Thus, the other cortical stratified transitional fields (STFs 2–6) could not be distinguished in the smaller fetuses. The early subplate contained axon bundles which were strongly positive for GAP43 (Fig. 1F). Calretinin-positive cells were almost evenly distributed throughout all layers of the neocortex (Fig. 3B). In the larger fetuses, the subplate or STF 1, the superficial fibrous layer or putative subcortical white matter, was characterized by (1) diffuse expression of HA (Fig. 2F), (2) strong expression of GAP 43 (Fig. 2B), (3) few GFAP-positive fibers (Fig. 2G) and (4) absence of PCNA-positive cells (Fig. 2D). If, in larger fetuses, STF 5, consisting of the deep cellular layer or the first transitional zone to appear outside the germinal matrix, was defined as a thin layer containing HA-bound cells in AEC reaction (Fig. 2F), then STF 2–4 and 6 could be identified based on the different arrangements of GFAP-positive fibers between layers (Fig. 2G). Nevertheless, DAB reaction did not demonstrate HA-positive cells in the probable STF 5 despite the strong positivity in the layer 1 and subplate (Fig. 3A). Moreover, even with AEC reaction, the positive cells in the STF5 were not found in our later trial 6 months after the sectioning (figures, not shown). Calretinin-positive cells were rarely seen and, if present, mostly distributed in the STF 2–6 or the intermediate zone (Fig. 3C). Laminar expression of HA was not observed in the large olfactory bulb, the thalamus, the ganglionic eminence (the putative basal ganglia) or the brainstem of both larger and smaller fetuses (Fig. 4). In sites other than the neocortex, HA-positive cells were distributed widely and evenly; we were unable to identify a specific topographical relationship between the putative nucleus and HA expression.
426
S. Shibata et al. / Annals of Anatomy 195 (2013) 424–430
Fig. 1. Expression of hyaluronan and intermediate filaments in the neocortex (putative temporal lobe) of a 12-week human fetus. Nearly horizontal sections. Panels A–F display proximate sections. Hyaluronan (panel A) is colored brown with diaminobenzidine (DAB) reaction. Panel B, glial fibrillary acidic protein (GFAP); panel C, vimentin; panel D, nestin; panel E, proliferating cell nuclear antigen (PCNA); panel F, growth associated protein-43 (GAP 43). All panels are presented at the same magnification (scale bar in panel C). Hyaluronan expression is seen in the layer 1 (or the marginal zone), the cortical plate and the early subplate (subplate). NEP, neuroepithelium; SVZ, subventricular zone.
The developing pyramidal tract was negative for HA (not shown). However, at the stages examined, the hippocampus was not well differentiated, but appeared as a simple fold of the cortex. The morphology of the cerebellum was also primitive, without laminar architecture or folds. 4. Discussion Using an anatomic atlas (Bayer and Altman, 2005), we attempted to identify cortical STFs in mid-term fetuses. The cortical plate was easily identified by its high cell density, whereas the subplate was characterized by the specific results of immunostaining for HA, GAP43, and PCNA. Our identified subplate seemed to correspond to that in Bayatti et al. (2008) and Ip et al. (2011) because of the strong expression of GAP43. However, the differentiation was still in a primitive stage because of few calretinin-positive neurons (present study) as well as because of no or few
nNOS-positive neurons (unpublished data), both of them are dominant populations of the definite subplate (Downen et al., 1999; Ulfig et al., 2000) at 12–16 weeks. Bayatti et al. (2008) also demonstrated a strong expression of synaptophysin in the subplate, but we failed to stain it in the present materials (unpublished data). Although the present subplate corresponded to STF 1, it was difficult to clearly discriminate between STFs 2–6, especially in small fetuses at 12–13 weeks. These layers seemed to correspond to the intermediate zone in Ulfig et al. (2000), Bystron et al. (2008) and Judaˇs et al. (2010). However, strangely, in the ventricular side of the subplate or STF 1, PCNA-positive cells were distributed evenly in the intermediate zone as well as in the ventricular and subventricular zone. In large fetuses at 15–16 weeks, arrangement of GFAP-positive fibers as well as laminar expression of HA suggested possible demarcations between these STFs: our hypothetically identified STFs are shown in Fig. 2G. The subplate is known to express fibronectin, chondroitin-sulfate proteoglycans,
S. Shibata et al. / Annals of Anatomy 195 (2013) 424–430
427
Fig. 2. Expression of hyaluronan and intermediate filaments in the neocortex (putative temporal lobe) of a 16-week human fetus. Sagittal sections. Panels A–E exhibit proximate sections. Hyaluronan (panels A and F) is stained red with the 3-amino-9-ethylcarbazole (AEC) reaction. Panel B, growth associated protein-43 (GAP 43); panel C, vimentin; panel D, proliferating cell nuclear antigen (PCNA); panel E, nestin; panel G, glial fibrillary acidic protein (GFAP). Panel F displays a higher magnification view of panel A. Panels A–E (or panels F and G) are presented at the same magnification (scale bar in panel A or G). Hyaluronan expression is seen in the layer 1 (or the marginal zone), the early subplate (subplate) and some deep cells near the subventricular zone (SVZ). NEP, neuroepithelium.
neural cell adhesion molecule (NCAM) and laminin (reviewed by Ulfig et al., 2000). To our knowledge, this study is the first to show laminar or stratified expression of HA in the human fetal neocortex. Notably, laminar expression was restricted to the neocortex, being absent from other parts of the brain in fetuses of 12–16 weeks. Thus, HA was likely to be committed to form an inside-out-gradient in the neocortex. Despite numerous studies on perineuronal nets, there have been few reports of laminar expression of proteoglycans and none of HA. HA was first expressed in superficially migrating cells,
followed by expression in the cortical plate at 12–13 weeks and later in the most superficial layer (layer 1) as well as in putative subcortical white matter (STF 1 or the subplate). The HA accumulation seemed to occur in accordance with the cortical plate formation. This result was consistent with findings showing that HA-binding glycoprotein was expressed in the cerebral cortex, specifically in the intermediate layer, of fetal rats at 15–19 days of gestation, but was not expressed in the cortical plate (Delpech and Delpech, 1984). The cortical plate cells seemed to produce HA in the STF 1 during their superficial migration to harbor axonal plexus of various
428
S. Shibata et al. / Annals of Anatomy 195 (2013) 424–430
Fig. 3. Expression of hyaluronan after diaminobenzidine (DAB) reaction and distribution of calretinin-positive cells. Sagittal sections. Panel A (16 weeks) displays hyaluronan expression in the neocortex after diaminobenzidine (DAB) reaction. No positive cells are seen in the intermediate zone between the subplate and the subventricular zone (SVZ). Panel B (12 weeks) and panel C (16 weeks) exhibit distribution of calretinin-positive cells. Calretinin-positive cells are few in number in the larger specimen.
afferent systems, while Cajal-Retzius cells were likely to accumulate HA in the layer 1. In addition to the tangential migration, HA-positive cells in the intermediate zone (STF-5 in the present figure) were also likely to be cooperated into the subplate. This possibility of axial migration seems to be consistent with Wang et al. (2010). The HA-rich zones, which sandwiched the cortical plate, might prevent further migration of cortical cells as HA-associated adhesive cues in the hippocampus (Förster et al., 2001). The rich HA in the subplate seemed to induce cell proliferation and differentiation as the other proteoglycans do (Li et al., 2005; Girós et al., 2007). In contrast to the fibrous or amorphous structures of HA in layer 1 and STF 1, spotty expression of HA in the deep layer surrounding or bound to some cells enabled us to identify the deep layer as STF 5. These HA-positive deep cells were interesting, but we did not find them after DAB reaction. Moreover, even with 3-amino-9-ethylcarbazole (AEC) reaction, we failed to stain them
on the old sections 6 months after preparation. Thus, they were likely to be a result of individual difference in development or, simply, an artifact in AEC reaction for an unknown reason. Perineuronal nets are composed of extracellular matrix molecules, including chondroitin sulfate proteoglycans, HA, tenascin-R, and link proteins, which are interposed between glial processes and nerve-cell surfaces in adult brains (Carulli et al., 2006). We did not examine the expression of chondroitin sulfate proteoglycans, but laminar expression of HA was most likely different from HA expression in perineuronal nets. Fiber-like HA structures in the putative white matter of mouse cerebellum were found to develop postnatally, with these structures being in the vicinity of inhibitory interneuron precursors and Purkinje cells (Baier et al., 2007). Although these structures may be regarded as distinct from perineuronal nets, the pattern of HA expression in the cerebellum does not indicate a laminar architecture but a
S. Shibata et al. / Annals of Anatomy 195 (2013) 424–430
429
Fig. 4. Expression of hyaluronan and intermediate filaments in the brain stem of a 12-week human fetus. Nearly horizontal sections. Panels A–C display neighboring sections. Hyaluronan (panel A) is colored brown with diaminobenzidine (DAB) reaction. Panel B, glial fibrillary acidic protein (GFAP); panel C, nestin. All panels are presented at the same magnification (scale bar in panel C). Hyaluronan-positive cells are distributed widely in the section. Nestin-positive fibers are evident from the interpeduncular area to the putative septal nucleus complex (panel C). III, putative oculomotor nucleus complex.
meshwork evenly distributed throughout the putative white matter. Thus, these HA structures seemed different from HA expression in STF 1. Actually, we found that the developing pyramidal tract was negative for HA. Laminar expression of HA in the superficial and deep sides of the cortical plate may result from the differentiation of cortical neurons, inducing a layer of putative white matter. Acknowledgment This study was supported by a grant (0620220-1) from the National R & D Program for Cancer Control, Ministry of Health & Welfare, Republic of Korea.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.aanat. 2013.07.002.
References Baier, C., Baader, S.L., Jankowski, J., Gieselmann, V., Schilling, K., Rauch, U., Kappler, J., 2007. Hyaluronan is organized into fiber-like structures along migratory pathways in the developing mouse cerebellum. Matrix Biol. 26, 348–358. Bayatti, N., Moss, J.A., Sun, L., Ambrose, P., Ward, J.F., Lindsay, S., Clowry, G.J., 2008. A molecular neuroanatomical study of the developing human neocortex from 8
430
S. Shibata et al. / Annals of Anatomy 195 (2013) 424–430
to 17 postconceptional weeks revealing the early differentiation of the subplate and subventricular zone. Cereb. Cortex 18, 1536–1548. Bayer, A.H., Altman, J., 2005. The human brain during the second trimester. In: Bayer, A.H., Altman, J. (Eds.), Atlas of Human Central Nervous System Development Series, vol. 3. CRC Press of Taylor & Francis group, Boca Raton, France, p. 34, 65, 78, 79 (temporal lobe); pp. 100–109 (brain stem). Bignami, A., Hosley, M., Dahl, D., 1993. Hyaluronic acid and hyaluronic acid-binding proteins in brain extracellular matrix. Anat. Embryol., 419–433. Bystron, I., Blakemore, C., Rakic, P., 2008. Development of the human cerebral cortex: Boulder Committee revisited. Nat. Rev. Neurosci. 9, 110–122. Carulli, D., Rhodes, K.E., Brown, D.J., Bonnert, T.P., Pollack, S.J., Oliver, K., Strata, P., Fawcett, J.W., 2006. Composition of perineuronal nets in the adult rat cerebellum and the cellular origin of their components. J. Comp. Neurol. 494, 559–577. Clowry, G., Molnár, Z., Rakic, P., 2010. Renewed focus on the developing human neocortex. J. Anat. 217, 276–288. Delpech, A., Delpech, B., 1984. Expression of hyaluronic acid-binding glycoprotein, hyaluronectin, in the developing rat embryos. Dev. Biol. 101, 391–400. Downen, M., Zhao, M.L., Lee, P., Weidenheim, K.M., Dickson, D.W., Lee, S.C., 1999. Neuronal nitric oxide synthase expression in developing and adult human CNS. J. Neuropathol. Exp. Neurol. 58, 12–21. Förster, E., Zhao, S., Frotscher, M., 2001. Hyaluronan-associated adhesive cues control fiber segregation in the hippocampus. Development 128, 3029–3039. Frischknecht, R., Seidenbecher, C.I., 2008. The crosstalk of hyaluronan-based extracellular matrix and synapses. Neuron Glia Biol. 4, 249–257. Giamanco, K.A., Matthews, R.T., 2012. Deconstructing the perineuronal net: cellular contributions and molecular composition of the neuronal extracellular matrix. Neuroscience 218, 367–384. Girós, A., Morante, J., Gil-Sanz, C., Fairén, A., Costell, M., 2007. Perlecan controls neurogenesis in the developing telencephalon. BMC Dev. Biol. 7, 29.
Ip, B.K., Bayatti, N., Howard, N.J., Lindsay, S., Clowry, G.J., 2011. The corticofugal neuron-associated genes ROBO1, SRGAP1 and CTIP2 exhibit an anterior to posterior gradient of expression in early fetal human neocortex development. Cereb. Cortex 21, 1395–1407. ˙ N., 2010. Populations of subJudaˇs, M., Sedmak, G., Pletikos, M., Jovanov-Miloˇsevic, plate and interstitial neurons in fetal and adult human telencephalon. J. Anat. 217, 381–399. Kwok, J.C., Dick, G., Wang, D., Fawcett, J.W., 2011. Extracellular matrix and perineuronal nets in CNS repair. Dev. Neurobiol. 71, 1073–1089. Lander, C., Kind, P., Maleski, M., Hockfield, S., 1997. A family of activity-dependent neuronal cell-surface chondroitin sulfate proteoglycans in cat visual cortex. J. Neurosci. 17, 1928–1939. Li, H.P., Oohira, A., Ogawa, M., Kawanura, K., Kawano, H., 2005. Aberrant trajectory of thalamocortical axons associated with abnormal localization of neurocan immunoreactivity in the cerebral neocortex of reeler mutant mice. Eur. J. Neurosci. 22, 2689–2696. Meyer, G., Schaaps, J.P., Moreau, L., Goffinet, A.M., 2000. Embryonic and early fetal development of the human neocortex. J. Neurosci. 20, 1858–1868. Shibata, S., Fukada, K., Imai, H., Abe, T., Yamashita, Y., 2003. In situ hybridization and immunohistochemistry of versican, aggrecan, and link protein and histochemistry of hyaluronan in the developing mouse limb bud cartilage. J. Anat. 203, 425–432. Ulfig, N., Neudörfer, F., Bohl, J., 2000. Transient structures of the human fetal brain: subplate, thalamic reticular complex, ganglionic eminence. Histol. Histopathol. 15, 771–790. Wang, W.Z., Hoerder-Suabedissen, A., Oeschger, F.M., Bayatti, N., Ip, B.K., Lindsay, S., Supramaniam, V., Srinivasan, L., Rutherford, M., Møllgård, K., Clowry, G.J., Molnár, Z., 2010. Subplate in the developing cortex of mouse and human. J. Anat. 217, 368–380.