COMMENT THIS ARTICLE
SEE COMMENTS ON THIS ARTICLE
CONTACT AUTHOR
December 1998 (Volume 39, Number 4)
Fibronectin Expression in the Developing Human Spinal Cord, Nerves, and Ganglia Mirko Krolo, Katarina Viloviæ1, Damir Sapunar, Eduard Vrdoljak2, Mirna Saraga-Babiæ Department of Histology and Embryology; 1Department of Anatomy, Split University School of Medicine; and 2Center for Oncology and Radiotherapy, Split University Hospital, Split, Croatia Aim. Analysis of developmental role of fibronectin during differentiation of the human spinal cord, nerves, and ganglia. Methods. Seven normal human embryos and fetuses between the 7th and 9th developmental week and a 9-week fetus with cervical spina bifida were histologically examined on hematoxylin and eosin stained serial paraffin sections of thoracic axial segments. Monoclonal antibody to the human cell fibronectin fragment was used for immunohistochemical detection of fibronectin. Results. In the 7th and 8th week of development, fibronectin was weakly expressed in the ventricular and intermediate zones of the spinal cord. Intense fibrillar expression was found in the marginal zone of the spinal cord – first over the ventral gray horns and later over the lateral and dorsal gray horns, and along the pathways of ventral and dorsal roots of the spinal nerves and in the spinal ganglia. At 9th week, fibronectin expression disappeared in the ventricular and intermediate zones and became weak and granular in the marginal zone of the spinal cord. In the spinal cord of a 9-week malformed fetus with cervical spina bifida, fibronectin expression was completely absent. Fibronectin was expressed in the nerves and ganglia throughout the investigated period, both in normal and malformed human conceptuses. Conclusion. Transient expression of fibronectin in the human spinal cord coincided with the most intense neuronal differentiation. Temporal and spatial expression of fibronectin during normal development, and its absence in a malformed human fetus suggests developmental role of fibronectin for the normal formation of the spinal cord. Key words: central nervous system; extracellular matrix; fibronectins; human development; immunohistochemistry; spinal cord; spinal dysraphism; spinal nerves Fibronectin is a cell-surface associated glyco- protein and one of the major components of the extracellular matrix. Fibronectin contains three types of sequence homology (type I, II, and III) organized into structural domains with interaction sites for other extracellular matrix molecules and cell surface glycoproteins (1-4). During development, components of the extracellular matrix seem to modulate cell differentiation and migration (5-7). Fibronectin is first found between the cells of the inner cell mass of the late mouse blastocyst (8), and prior to streak formation in the chick (9). However, the developmental role of fibronectin is mostly associated with morphogenetic movements during gastrulation (10,11). After gastrulation, fibronectin is broadly distributed in extracellular spaces of the connective tissue and basement membranes of epithelia. Fibronectin probably mediates both anchorage of fibroblasts and locomotion of some embryonic cells since it is required for cell adhesion and motility both in vivo and in vitro (11). During neural tube formation, fibronectin is found along the pathways of migration of the neural crest cells (11). Experimentally, fibronectin stimulates neurite extension of the chick embryonic central neurons in the cell culture (12). In early human development, the neural tube forms during the 4th week and later differentiates into the brain and spinal cord. The cranial part of the neural tube forms during the process of primary neurulation, whereas its most caudal part differentiates during secondary neurulation (13-16). Neuroepithelial cells in the ventricular zone subsequently differentiate into primitive neurons – neuroblasts. Neuroblasts become neurons as they develop cytoplasmic processes which constitute the marginal zone (17). The elongation of neurites may be promoted by regional and developmetal differences in the distribution of extracellular matrix components, such as laminin and fibronectin. In the adult human nervous tissue, fibronectin is found only in the walls of the brain capillary vessels and in the basement membranes of perineurial Schwann cells of the peripheral nerves (1). Experimentally, fibronectin was not associated with the differentiation of neurons and glia in many areas of the central nervous system; transient fibronectin expression in the mouse was found only in
the developing cerebellum (12,18). Data on the possible role of fibronectin during development of the human nervous system are not available. To elucidate the possible role of fibronectin during the differentiation of the human nervous system, we histologically examined the appearance and distribution of fibronectin in the spinal cord of normal human embryos between the 7th the and 9th week of development, and compared it to the fibronectin expression in a 9-week human fetus with cervical spina bifida. Material and Methods Seven normal human embryos and fetuses between the 7th and 9th developmental week, and a 9week human fetus with cervical spina bifida were collected after spontaneous or legal abortions at the Department of Gynecology and Obstetrics, Split University Hospital. Human tissue was treated as postmortal material with permission of the hospital’s Drug and Ethical Committee and in accordance with the Helsinki Declaration. The postovulatory age was estimated from the menstrual data or ultrasonography appearance and correlated with the crown-rump length and Carnegie stages (19). Thoracic segments of axial structures (vertebral column and spinal cord) were dissected from embryos and fetuses, fixed in 4% paraformaldehyde in phosphate buffer for several hours and embedded in paraffin. Tissue blocks were serially cut in a transversal direction and mounted on glass slides coated with 0.1% poly-L-lyzine. Morphological analysis was performed by staining the 7 mm sections with hematoxylin and eosin. For immuno- histochemical detection of fibronectin, a purified monoclonal antibody to the human cell fibronectin fragment (Boehringer, Mannheim, Germany) was used. Incubation with primary antibody (20 g/mL in phosphate buffered saline, PBS) was performed for 1 hour at room temperature. Peroxidase quenching was performed with 30% hydrogen peroxide dissolved in absolute methanol. The primary antibody was visualized with Histostain SP-Peroxidase kit (Zymed Laboratorie Inc., San Francisco, Calif, USA) and diaminobenzidine. The sections were counterstained with hematoxylin. Sections stained without the primary antibody were used as controls. Results In the 7th week of development, the wall of the human spinal cord consisted of three zones: ventricular, intermediate, and marginal zone. At that stage, the well developed ventral horn of the gray matter consisted of motor neuroblasts whose extended neurites formed ventral roots of the spinal nerves. The dorsal horns received centrally directed neurons (dorsal roots) of neural crest cells in the dorsal root ganglia. A very weak fibronectin expression characterized only the dorsal part of ventricular zone, whereas a fine network of fibronectin fibrils was observed among the neuroblasts in the intermediate zone. Strong expression of fibro- nectin fibrils was seen throughout the marginal zone over the ventral horns and medial (inner) parts of lateral and dorsal horns (Figs. 1a,b). Outer (surface) parts of the marginal zone of these two horns, as well as the floor plate area, did not stain. Fibronectin fibrils were found on the surface and inside the spinal ganglia, as well as within the spinal nerves in parallel strands along their pathways (Fig. 1c). Fine fibronectin network was seen in the connective tissues of different organ systems (data not shown). Figure 1a: Fibronectin immunostaining of the cross section through the thoracic axial organs of a 7-week human embryo. The wall of the spinal cord (sc) consists of the ventricular zone (v), the intermediate zone (i), and the marginal zone (m), floor plate (f), and roof plate (r). The ventral horns (vh) and dorsal horns (dh) of gray matter, spinal nerves (sn), and spinal ganglia (sg), as well as vertebral column (vc) are formed. Fibronectin expression (arrows) is seen in the marginal zone of the spinal cord and in the surrounding connective tissue. Hematoxylin and eosin, x80. [view this figure] Figure 1b: Detail of the Fig. 1a. Cross section through the dorso-lateral part of the 7-week human spinal cord: ventricular zone (v), intermediate zone (i), marginal zone (m). Expression of fibronectin (arrows) is visible in the intermediate and marginal zones (asterisk). Hematoxylin and eosin, x200. [view this figure] Figure 1c: Detail of the Fig. 1a. Cross section through the ventral half of a 7-week human spinal cord and surrounding structures: ventral horns (vh), floor plate (f), ventral roots of spinal nerves (sn), spinal ganglia (sg), connective tissue (c). Fibronectin fibrils (arrows) are seen within the intermediate zone (i) of ventral horns, and as parallel strands in the marginal zone (m), and spinal nerves. Hematoxylin and eosin, x250. [view this figure]
In the 8th week of development, the differentiation of the spinal cord progressed. The ventricular zone
was thinner, whereas the intermediate and marginal zones thickened due to the increasing differentiation of neurons and their processes (Fig. 2a). Very weak fibrillar fibronectin staining was occasionally found in the ventricular zone, mostly on the border with the intermediate zone. A fine network of fibronectin was still present in the intermediate zone. Strong expression of fibronectin fibrils was seen throughout the marginal zone over the ventral horns, lateral and dorsal horns, apart from the very thin surface layer. (Fig. 2b). Fibronectin fibrils characterized both spinal ganglia and spinal nerves (see Fig. 3c). Sections stained without the primary antibody were used as negative controls. In those sections neither fibrillar nor granular fibronectin expression was detected within the layers of the spinal cord, in the spinal nerves and ganglia. Fibronectin network was not visualized in the connective tissue of different organs. The spinal cord of negative controls had the same appearance as the sections stained with hematoxylin and eosin (see Fig.2a). Figure 2a: Cross section through the thoracic spinal cord of an 8-week human embryo, stained with hematoxylin and eosin: the ventricular zone (v) is thinner, while the intermediate (i) and marginal (m) zones thickened. Ventral (vh) and dorsal horns (dh) of gray matter, spinal nerves (sn), and ganglia (sg) are seen. Hematoxylin and eosin, x200. [view this figure] Figure 2b: Fibronectin immunostaining of the cross section through thethoracic spinal cord (sc) of an 8-week human embryo described in Fig. 2a. Fibronectin fibrils (arrows) have a parallel orientation throughout the marginal zone, except for the very thin surface layer across lateral and dorsal horns. The floor (f) and roof plates (r) do not express fibronectin. Hematoxylin and eosin, x220. [view this figure]
In the 9th week of development, the definitive spinal cord was formed: the neural canal shrank and became the central canal. The ventricular zone became the ciliated ependymal layer, and the ventral, lateral, and dorsal funiculi of the white matter were formed by ascending and descending pathways of neurons. Fibronectin fibrils were not present in the ependymal and intermediate layers. In the marginal layer, the expression and distribution of fibronectin changed compared to the earlier stages: it was very faint and lost its parallel orientation (Fig. 3a). Very weak granular expression of fibronectin was observed throughout the marginal layer (Fig. 3b). Fibronectin fibrils were observed in the spinal ganglia and along the pathways of the spinal nerves, and in the connective tissue of different organs (Fig. 3c). Figure 3a: Fibronectin immunostaining of the cross section through thethoracic spinal cord (sc), spinal nerves (sn), and spinal ganglia (sg) of a 9-week human fetus. The spinal cord consists of the central canal (cc) surrounded by the ependymal cells (e), ventral horns (vh), and dosal horns (dh) of gray matter, and the marginal zone (m) which transformed into the corresponding funiculi. Weak granular fibronectin expression (arrows) is seen throughout the marginal zone. Hematoxylin and eosin, x200. [view this figure] Figure 3b: Detail of the Fig.3a. Granular expression of fibronectin (arrows) in the marginal layer (m) of the dorsal funiculi overlying the dorsal horns (dh). Hematoxylin and eosin, x250. [view this figure] Figure 3c: Detail of the Fig. 3a. Parallel strands of fibro- nectin fibrils (arrows) characterize spinal nerves (sn) of a 9-week fetus. Hematoxylin and eosin, x250. [view this figure]
In the 9-week fetus with cervical spina bifida, the thoracic part of the spinal cord displayed changes in the organization of cell layers: the central canal and areas of floor plate, and roof plate were irregular. The marginal layer was thin and irregular, particularly in the region of the ventral funiculi (Fig. 4a). We could not observe either fibrillar or granular fibronectin expression in the layers of the spinal cord of the malformed fetus, although fibrillar fibronectin expression characterized the spinal nerves and ganglia, connective tissue of the surrounding meninges, and adjacent skin structures (Fig. 4b). Figure 4a: Fibronectin immunostaining of the cross section through the thoracic axial organs of a 9week human fetus with cervical spina bifida. The spinal cord (sc) has an an irregular central canal (cc), floor (f), and roof plates (r). The marginal layer (m) in the region of ventral funiculi is of unequal thickness. Ventral horns (vh) and dorsal horns (dh) of gray matter, spinal ganglia (sg), vertebral column (vc), vertebral arches (va) are seen. Fibronectin is not expressed within the spinal cord. Hematoxylin and eosin, x50. [view
this figure] Figure 4b: Detail of Fig. 4a. Cross section through the part of ventral horns (vh), irregular marginal layer (m) of ventral funiculi, floor plate (f), and connective tissue (c). FN is not expressed in the layers of the spinal cord, although fibronectin fibrils (arrows) are seen in the surrounding connective tissue. Hematoxylin and eosin, x200. [view this figure]
Discussion Our investigation demonstrated transient expression of fibronectin during the formation of the human spinal cord. During embryonic period of development (7th and 8th developmental weeks), intense differentiation of the spinal cord was associated with the expression of fibronectin within all the three layers of the spinal cord. The expression was particularly intense in the marginal layer, formed of citoplasmic processes of differentiating neurons. Such findings indicate that fibronectin could participate in the elongation of neurits of the human spinal cord neurons. In the early fetal period (9th week), cessation of the differentiation of neurons in the lateral walls of the spinal cord coincided with the reduction of fibronectin expression: first in the ventricular and intermediate zones, and then in the marginal zone. It seems that, after the onset of spinal cord differentiation, fibronectin is not required for further steps of its development. Transient presence of fibronectin in the layers of the developing human spinal cord has not been described before. It was found only in the mouse developing cerebellum (12,18). In our study, fibronectin fibrils were also found within the developing peripheral nerves in the dorsal roots (derived from the neural crest cells) and ventral roots (derived from neurons in the ventral horns) of the spinal nerves. Therefore, fibronectin seems to participate in the formation of the spinal nerves and guidance of their neurites to the final position in the target tissue. This supports the experimental data that the migration of embryonic and regenerating neurits along long distances is guided by extracellular matrix components (20). In vitro, the rate of neurite elongation and guidance of growth cones migration can be influenced by the adhesivity of the substratum (21). However, central and peripheral neurons differ in their response to fibronectin, which can be explained by the difference in their cell surface antigens (11,12). Previous reports on the developing human nervous tissue showed fibronectin expression only at the sites of contact between the neuro- ectoderm and mesenchyme. Investigations on adult human tissues described fibronectin expression in the basement membranes of perineurial and Schwann cells and at the Ranvier nodes (1). Fibronectin was not found either in neurons or glial cells in the human brain but was found only in the walls of capillaries (1). During embryogenesis, fibronectin expression was closely linked to cell differentiation and organogenesis (5-8). Expression of fibronectin was also found in the differentiating human notochord (13). Similar to the experimental data, fibronectin first appeared in the basement membrane, but later also along the cell surface of migrating notochord cells (13,22). In the 9th week of the fetal development, cessation of neuronal differentiation was associated with a decrease of fibronectin expression in the spinal cord. In the malformed human fetus of the same developmental age with the cervical spinal bifida fibronectin was absent in the spinal cord. This suggests the importance of temporal and spatial expression of fibronectin for the normal formation of the spinal cord. Our study on fibronectin expression in the developing human nervous tissue presents a starting point for further analysis of its role during human embryogenesis. Acknowledgments We thank Ms Asja Miletiæ for her skillful technical assistance. This work was supported by the Croatian Ministry of Science and Technology, Grant No. 108-194. References 1 Paetau A, Mellstrom K, Vaheri A, Haltia M. Distribution of a major connective tissue protein, fibronectin, in normal and neoplastic human nervous tissue. Acta Neuropathol 1986;51:47-51. 2 Mosher DF, Furcht LT. Fibronectin: rewiev of its structure and possible functions. J Invest Dermatol 1981;77:175-80. 3 Ruoslahti E, Engvall E, Hayman EG. Fibronectin: current concepts of its structure and function. Coll Res 1981;1:95-128. 4 Darribere T, Koteliansky VEK, Chernousov MA, Akiyama SK, Yamada KM, Thiery JP, et al. Distinct regions of human fibronectin are essential for fibril assembly in an in vivo developing system. Dev Dyn 1992;194:63-70. 5 Lofberg J, Ahlfors K, Fallstrom C. Neural crest cell migration in relation to extracellular matrix organization in the embryonic axolotl trunk. Dev Biol 1980;75:148-67. 6 Mayer BW, Hay ED, Hynes RO. Immuno- cytochemical localization of fibronectin in embryonic chick
trunk and area vasculosa. Dev Biol 1981;82:267-86. 7 Sugrue SP, Hay ED. Interaction of embryonic corneal epithelium with exogenous collagen, laminin, and fibronectin: role of endogenous protein sythesis. Dev Biol 1982;92:97-106. 8 Wartiovaara J, Leivo I, Vaheri A. Expression of the cell surface-associated glycoprotein, fibronectin, in the early mouse embryo. Dev Biol 1979;69:247-59. 9 Mitrani E, Farberov A. Fibronectin expression during the processes leading to axis formation in the chick embryo. Dev Biol 1982;91:197-201. 10 Boucaut J-C, Darribere T, Boulekbache H, Thiery JP. Prevention of gastrulation but not neurulation by antibodies to fibronectin in amphibian embryos. Nature 1984;307:364-67. 11 Dufour S, Duband J-L, Kornblihtt AR, Thiery J-P. The role of fibronectins in embryonic cell migrations. Trends Genet 1988;4:198-203. 12 Rogers SL, Letourneau PC, Palm SL, McCarthy J, Furcht LT. Neurite extension by peripheral and central nervous system neurons in response to substratum-bound fibronectin and laminin. Dev Biol 1983;98:212-20. 13 Saraga-Babiæ M, Lehtonen E, Švajger A, Wartiovaara J. Morphological and immunohistochemical characteristics of axial structures in the transitory human tail. Ann Anat 1994;176:277-86. 14 Saraga-Babic M, Sapunar D, Wartiovaara J. Variations in the formation of the human caudal spinal cord. Anat Anzeiger 1995;36:341-7. 15 Saraga-Babic M, Krolo M, Sapunar D, Terzic J, Biocic M. Differences in origin and fate between the cranial and caudal spinal cord during normal and disturbed human development. Acta Neuropathol 1996;91:194-9. 16 Saraga-Babiæ M, Stefanoviæ V, Lehtonen E, Sapunar D, Saraga M, Wartiovaara J. Neurulation mechanisms in the human development. Croatian Med J 1996;37:7-14. 17 Moore KL, Persaud TVN. The nervous system. In: McGrew L, Kilmer L, editors. The developing human. Clinically oriented embryology. Philadelphia: Sounders; 1993. p. 385-422. 18 Hatten ME, Furie MB, Rifkin DB. Binding of developing mouse cerebellar cells to fibronectin: a possible mechanism for the formation of the external granular layer. J Neurosci 1982;2:1195-206. 19 O’Rahilly R, Gardner R. The timing and sequence of events in the development of the human nervous system during the embryonic period proper. Z Anat Entwickl Gesch 1971;134:1-12. 20 Bohn RC, Reier PJ, Sourbeer EB. Axonal interactions with connective tissue and glial substrata during optic nerve regeneration in Xenopus larve and adults. Am J Anat 1982;165:397-419. 21 Letourneau PC. Cell-to-substratum adhesion and guidance of axonal elongation. Dev Biol 1975;44:92-102. 22 Harrison F, Van Nassauw L, Van Hoof J, Foidart J-M. Microinjection of antifibronectin antibodies in the chicken blastoderm: inhibition of mesoblast cell migration but not of cell ingression at the primitive streak. Anat Rec 1993;236:685-96. Received: July 22, 1998 Accepted: August 25, 1998 Correspondence to: Mirna Saraga-Babiæ Department of Histology and Embryology, Split University School of Medicine, PAK, KB Split Spinèiæeva 1 21000 Split, Croatia
[email protected]
Copyright © 1997 by the Croatian Medical Journal. All rights reserved. Created 16/1/99 - Last Modified 16/1/99 Created and maintained by: Tinman
