10
J. Embryol. exp. Morph. 76, 27-36 (1983) 27 Printed in Great Britain © The Company of Biologists Limited 1983 Vascular architecture of the developing spinal cord in the rat: a suggested model By R. SIMON-MARIN 1 , J. R. VILANOVA, A. AGUINAGALDE AND E. BARBERA-GUILLEM 2 From the Department of Biology and Histology, University of the Basque Country, Bilbao, Spain SUMMARY The vascular architecture of the developing Sprague-Dawley rat spinal cord from E l l through E16 is reported. The paraffin-embedded cord is serially sectioned in the sagittal, transversal and frontal planes and stained with PAS and methenamine silver. Serial semithin transverse sections are stained with toluidine blue. The results demonstrate two highly integrated vascular systems: one sagittally disposed in three concentric networks and the other radially oriented around the cord. The sagittal plexus is configurated by rhombohexagonal polygons. The lateral radial stem vessels anastomose with the sagittal systems at the polygonal vertex. A structural vascular model of the cord is proposed. The periodical sequence distribution of vessels in the three planes and their relationship to spinal ganglia is suggestive of a neural metamera vascularly determined. INTRODUCTION Intrinsic CNS vascular embryogenesis is still poorly understood at a descrip- tive fragmentary level. Light microscopy observations mainly formulated on ink- coloured thick sections from birds and mammalian brains and cords have established the presence of an early deep subependymal plexus in connection with a perineural pial net through a system of transversally penetrating vessels (Lewis, 1957; Strong, 1947, 1961, 1964; Sims, 1961; Roncali, Camosso & Am- brosi, 1973; Camosso, Roncaly & Ambrosi, 1973, 1976). The penetrating radial arrangement of developing blood vessels in the mam- malian brain has been well documented by microangiography (Stoeter, Schmidt- Lademann & Voigt, 1978). Similar findings consisting of a plexiform internal plexus and a set of penetrat- ing stem vessels have been reported in the telencephalic wall of human embryos and foetuses when treated with the Golgi and reticulin stains (Duckett, 1971). Ultrastructural studies dealing with CNS vascular embryogenesis in birds and 1 Author's address: Departamento de Anatomia, Hospital de la Seguridad Social Cruces, Baracaldo, Spain. 2 Author's address: Departamento de Biologia Celular e Histologia, Universidad del Pais Vasco, Lejona, Bilbao, Vizcaya, Spain.

Vascular architecture of the developing spinal cord in the rat ...Developing spinal cord in the rat 33 penetration is ventral to the dorsal root and the third and fourth occupy the

  • Upload
    others

  • View
    3

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Vascular architecture of the developing spinal cord in the rat ...Developing spinal cord in the rat 33 penetration is ventral to the dorsal root and the third and fourth occupy the

J. Embryol. exp. Morph. 76, 27-36 (1983) 2 7Printed in Great Britain © The Company of Biologists Limited 1983

Vascular architecture of the developing spinal cordin the rat: a suggested model

By R. SIMON-MARIN1, J. R. VILANOVA, A.AGUINAGALDE AND E. BARBERA-GUILLEM2

From the Department of Biology and Histology, University of the BasqueCountry, Bilbao, Spain

SUMMARY

The vascular architecture of the developing Sprague-Dawley rat spinal cord from El lthrough E16 is reported. The paraffin-embedded cord is serially sectioned in the sagittal,transversal and frontal planes and stained with PAS and methenamine silver. Serial semithintransverse sections are stained with toluidine blue. The results demonstrate two highlyintegrated vascular systems: one sagittally disposed in three concentric networks and the otherradially oriented around the cord. The sagittal plexus is configurated by rhombohexagonalpolygons. The lateral radial stem vessels anastomose with the sagittal systems at the polygonalvertex. A structural vascular model of the cord is proposed. The periodical sequencedistribution of vessels in the three planes and their relationship to spinal ganglia is suggestiveof a neural metamera vascularly determined.

INTRODUCTION

Intrinsic CNS vascular embryogenesis is still poorly understood at a descrip-tive fragmentary level. Light microscopy observations mainly formulated on ink-coloured thick sections from birds and mammalian brains and cords haveestablished the presence of an early deep subependymal plexus in connectionwith a perineural pial net through a system of transversally penetrating vessels(Lewis, 1957; Strong, 1947, 1961, 1964; Sims, 1961; Roncali, Camosso & Am-brosi, 1973; Camosso, Roncaly & Ambrosi, 1973, 1976).

The penetrating radial arrangement of developing blood vessels in the mam-malian brain has been well documented by microangiography (Stoeter, Schmidt-Lademann & Voigt, 1978).

Similar findings consisting of a plexiform internal plexus and a set of penetrat-ing stem vessels have been reported in the telencephalic wall of human embryosand foetuses when treated with the Golgi and reticulin stains (Duckett, 1971).

Ultrastructural studies dealing with CNS vascular embryogenesis in birds and

1 Author's address: Departamento de Anatomia, Hospital de la Seguridad Social Cruces,Baracaldo, Spain.

2 Author's address: Departamento de Biologia Celular e Histologia, Universidad del PaisVasco, Lejona, Bilbao, Vizcaya, Spain.

Page 2: Vascular architecture of the developing spinal cord in the rat ...Developing spinal cord in the rat 33 penetration is ventral to the dorsal root and the third and fourth occupy the

28 R. SIM6N-MARIN AND OTHERS

mammals are mainly restricted to in situ morphological angiogenetic phenomenaand are not concerned with mapping or the establishing of patterns (Phelps,1972; Roy, Hirano, Kochen & Zimmerman, 1974; Hauw, Berger & Escourolle,1975; Povlishock, Martinez & Moossy, 1977; Allsopp & Gamble, 1979).

The purpose of the present histological contribution is to report a highlydeveloped pattern of vascular morphology in the developing spinal cord of therat between the 12th and 16th days of embryonic life.

MATERIALS AND METHODS

Six sets of six Sprague-Dawley rat embryos from E l l through E16 were im-mersed in Bouin's solution and embedded in paraffin. The intravascularpresence of red blood cells was an excellent landmark to delineate vascularchannels. Two embryos from each gestational day were serially sectioned in thetransverse plane, two in the sagittal and two in the frontal plane. Half of theembryos obtained each day and from each section plane was routinely stainedwith the PAS method; the other half was stained with the Gomori silver-methenamine technique. An average of two hundred sections was obtained fromeach embryo spinal cord.

Representative 1 mm-thick transverse slices of the thoracic spinal cord fromE12 through E16 were fixed in glutaraldehyde and embedded in Araldite. Anaverage of 30 ljum-thick sections, obtained from each block, was routinelystained with toluidine blue.

RESULTS

Because of the variable grade of vascular development attained by the dif-ferent regions of the cord with a maximum of 2 days delay observed between thecervical and caudal segments, the following data refer to the thoracic region.

Ell

The six embryos studied revealed an avascular cord. The pial plexus is verypoorly developed ventrally and absent dorsally.

Fig. 1. Transversal plane. (A) E12 with ventral bilateral and symmetrical vascularpenetration. (B) E12 showing anastomoses between radial stem vessels andsubependymal plexus; arrow at ventral root. (C) E14 with first radial vessel coincid-ing with central root (arrow). (D) Semithin section from E13 displaying the 4th and5th radial vessels (arrows). (E & F) Right hemicord shows partial aspects of fourradial vessels (arrows); dotted line shows the course of the three concentric plexus.(A & B) PAS x 150. (C) Toluidine blue x 150. (D) Toluidine blue x 200. (E)Methenamine silver x 100. (F) Methenamine silver x 200.

Page 3: Vascular architecture of the developing spinal cord in the rat ...Developing spinal cord in the rat 33 penetration is ventral to the dorsal root and the third and fourth occupy the

Developing spinal cord in the rat 29

Fig. 1

EMB76

Page 4: Vascular architecture of the developing spinal cord in the rat ...Developing spinal cord in the rat 33 penetration is ventral to the dorsal root and the third and fourth occupy the

30 R. SIM6N-MARIN AND OTHERS

E12

Transversal plane

The vascular pial plexus is greatly developed on the ventral and ventrolateralaspects of the cord. The endothelial cells display a high number of mitoticfigures.

Bilaterally and symmetrically a large vessel enters the cord ventrally and nearthe midline (Fig. 1A). The vessel feeds a single sagittal deep subependymalplexus. A dorsal exit or communication with a practically absent dorsal pialvessels is not regularly seen on E12.

Two symmetrical lateral and radially oriented straight stem vessels penetratethe cord on the ventrolateral surface (Fig. IB). The first limits posteriorly theanterior horn; the second vessel enters the cord approximately midway betweenthe ventral and dorsal nerve roots. Both vessels anastomose with thesubependymal plexus (Fig. IB).

Sagittal plane

The ventral near midline symmetrical vessels enter the cord at equal intervals.The vessels fork shortly after entrance and run dorsally generating thesubependymal network. This deep sagittal plexus exhibits a polygonal ap-pearance.

Frontal plane

Sections confirm the anastomoses between the radial vessels and the sagittaldeep plexus.

E13

Transversal plane

The pial plexus has extended dorsally and endothelial cells are actively divid-ing there. The deep subependymal plexus has established connections at regularintervals with the dorsal pial vascular trama. The two radial lateral vesselspresent on E12 turn into five vascular penetrations (Fig. 1C, D, E). They aredistributed ventrodorsally as follows: the first coincides with the ventral rootarea, the second limits dorsally the ventral horn, the fifth and most dorsal lateral

Fig. 2. Sagittal and frontal planes. (A) Mid-sagittal view from E13 revealing deeppolygonal plexus. (B) Sagittal view of sequential ventral vessels from E13. (C)Ventral vessels, notice initiation of deep plexus, forking and rhombic shapes ventr-ally. (D) Close-up view of vascular entrance and forking on E13. (E) Frontal planeshowing relationship between stem radial vessels and spinal ganglia, E13. (A)Methenamine silver x 40. (B) PAS x 200. (C) Methenamine silver x 150. (E)PAS x 200.

Page 5: Vascular architecture of the developing spinal cord in the rat ...Developing spinal cord in the rat 33 penetration is ventral to the dorsal root and the third and fourth occupy the

Developing spinal cord in the rat

S

31

Fig. 2

Page 6: Vascular architecture of the developing spinal cord in the rat ...Developing spinal cord in the rat 33 penetration is ventral to the dorsal root and the third and fourth occupy the

32 R. SIM6N-MARIN AND OTHERS

Fig. 3

Page 7: Vascular architecture of the developing spinal cord in the rat ...Developing spinal cord in the rat 33 penetration is ventral to the dorsal root and the third and fourth occupy the

Developing spinal cord in the rat 33penetration is ventral to the dorsal root and the third and fourth occupy the spaceleft between the second and fifth vessels.

A second vascular network representing a duplicate of the deep subependymalplexus appears ventrally and lies on the pial side of the first. It runs a slightlydivergent course dorsally in relation to the ependymal layer. An anterior orposterior pial connection cannot be seen and is only sustained by the radialvessels.

Sagittal plane

The anterior vessels display a regular periodicity bifurcating deeply and sagit-tally; the upper and lower branches anastomose with the corresponding divisionsof the adjacent sequential vessels (Fig. 2C). These anastomoses represent themost ventral aspect of the deep plexus (Fig. 2C, D).

The geometrical pattern exhibited by the subependymal plexus, a polygonalrhombohexagonal architectural net (Figs 2A & 3C), confirms the low-powerobservations obtained from fresh mid-sagittal sections of whole embryos viewedthrough the ependymal layer (Fig. 3A, B). The capillary network shows a rhom-bic shape ventrally and an hexagonal tendency dorsally.

Frontal plane

The lateral vessels enter the cord in a sequentially ordered fashion. The frontalsections that are central to spinal ganglia show a three to one ratio relationshipbetween the vessels and the neural crest. Three vessels, one central and two polarcorrespond to one ganglion. The polar vessels often occupy the interganglionarspace (Fig. 2E).

El 4 and El 5

The only morphological event in relation to previous data is the appearanceof a third sagittal polygonal plexus. It also follows a ventrodorsal course andshows a similar tendency to diverge towards the posterior root. This third and lastappearing lateral vascular net is an image of the second and is only visible in themore mature regions of the cord (Fig. IE, F).

E16

The vascular architecture is obscured and no longer demonstrable by thismethod. The reasons are related to vascular area density. The neuropile hasgrown considerably and the vessels have become much thinner and straighter sothat they are less apparent on sections.

Fig. 3. (A & B) Low-power transependymal view from fresh mid-sagittal section;notice polygonal-shaped subependymal plexus. Right side of A is ventral andpredominantly rhombic, left tends to be hexagonal. (C) High-power sagittal viewdemonstrating polygonal net. (A) x200. (B) x250. (C) PAS x 400.

Page 8: Vascular architecture of the developing spinal cord in the rat ...Developing spinal cord in the rat 33 penetration is ventral to the dorsal root and the third and fourth occupy the

34 R. SIM6N-MARIN AND OTHERS

Fig. 4. The diagram depicts the proposed vascular model. The upper and innercylinder (1) includes the inner subependymal rhombohexagonal plexus with ventraland dorsal regularly sequential longitudinally running vessels. The second mid-uppercylinder (2) represents the second vascular network only sustained by radial stemvessels. The mid-lower third cyclinder (3) shows the external third concentricperipheral mesh. The lowest and most peripheral fourth cylinder (4) displays theentrances of the five lateral and radially oriented stem vessels.

Page 9: Vascular architecture of the developing spinal cord in the rat ...Developing spinal cord in the rat 33 penetration is ventral to the dorsal root and the third and fourth occupy the

Developing spinal cord in the rat 35

DISCUSSION

Analysis of data tends to demonstrate a highly organized vascular structurepresent in the spinal cord of the developing rat. The results can be reduced to twospatially determined systems that appear to provide the cord cylinder with anappropriate mesodermal skeleton (Fig. 4).

The first vascular system is sagittally oriented and parallel to the ependymalcanal. It comprises three concentric polygonal meshes. The inner one, seenthrough the ependyma is generated by the vessels which run ventrodorsally,enter near the midline and fork. The second and third appearing concentric nets,are closer to the pia and they seem to be a replica of the first (Fig. IF); they lacka ventral or dorsal connection.

The second system is represented by the lateral group of five radially orientedstem vessels. They penetrate deeply and end up anastomosing at the polygonalvertex of the three concentric sagittal nets (Fig. IF).

The orderly sequence of vascular entrances in the frontal and sagittal planes,the highly developed polygonal network generated by the ventral forking vesselsand the radial stem vessel-polygonal vertex anastomoses enables us to proposea vascular model structure of the cord illustrated in Fig. 4.

Previous data in relation to spinal cord vascularization is included in the tworepeatedly reported features: the penetrating radial stem vessels and the internalependymal plexus (Strong, 1961; Camosso etal. 1976). All of these findings areeasily incorporated into the model proposed here.

The rapid increase in vascularization reported in the mouse spinal cord betweenE14 and E17 (Sturrock, 1981) is basically similar to the phenomena observed inthe rat and it shows itself in the progressive configuration of the vascular rings.

More suggestive is the numerical ratio between the lateral penetrating vesselsand the spinal cord ganglia (Fig. 2E). The frontal sections indicate a suspiciouscoincidence of three vessels, one central and two polar with respect to oneganglion. This relates the metameric neural model to the proposed vascularmodel. The vascularization is obviously secondary to previous neural metamer-ism. It is likely that the vasculature contributes to metameric development andalthough it does not appear to be related to nervous tissue growth it may play arole in certain aspects of astroglial development.

A structural polyhedral unit limited by vessels may be inferred from themodel. It would be represented by a truncated rhombohexagonal pyramid radi-ally disposed in the cord cylinder. The smaller apical surface would coincide withthe corresponding subependymal polygon of the internal plexus, and the basedetermined by the entering radial vessels. The significance of this in relation toorganization of cell population in the nervous tissue still needs to be investigated.

The authors wish to thank Mrs Milagros Portuondo, Ms Cristina Otamendi and Mr JosebaGarcia Elizburu for their technical assistance, photography and illustrations.

Page 10: Vascular architecture of the developing spinal cord in the rat ...Developing spinal cord in the rat 33 penetration is ventral to the dorsal root and the third and fourth occupy the

36 R. SIM6N-MARIN AND OTHERS

REFERENCESALLSOPP, G. & GAMBLE, H. J. (1979). Light and electron microscopic observations on the

development of the blood vascular system of the human brain. /. Anat. 128, 461-477..CAMOSSO, M. E., RONCALI, L. & AMBROSI, G. (1973). Vascular patterns in the chick embryo

spinal cord. Proc. 3rd Eur. Anat. Congress, Manchester, 269-270.CAMOSSO, M. E., RONCALI, L. & AMBROSI, G. (1976). Vascular patterns in the chick embryo

spinal cord in normal and experimentally modified development. Acta anat. 95, 349-367.DUCKETT, S. (1971). The establishment of internal vascularization in the human

telencephalon. Acta anat. 80, 107-113.HAUW, J. J., BERGER, B. & ESCOUROLLE, R. (1975). Electron microscopic study of the

developing capillaries of the human brain. Acta Neuropath. 31, 229-242.LEWIS, O. J. (1957). The form and development of the blood vessels of the mammalian

cerebral cortex. /. Anat. 91, 40-46.PHELPS, C. H. (1972). The development of glio-vascular relationships in the rat spinal cord.

Z. Zellforsch. mikrosk. Anat. 128, 555-563.POVLISHOCK, J. T., MARTINEZ, A. J. & MOOSSY, J. (1977). The fine structure of blood vessels

of the telencephalic germinal matrix in the human fetus. Am. J. Anat. 149, 439-452.RONCALI, G., CAMOSSO, M. E. & AMBROSI, G. (1973). Schemi di vascolarizzazione del midollo

spinale deU'embrione di polio in condizioni normali. Boll. Soc. Ital. Biol. sper. 49,141-147.ROY, S., HIRANO, A., KOCHEN, J. A. & ZIMMERMAN, H. M. (1974). The fine structure of

cerebral blood vessels in chick embryo. Acta Neuropath. 30, 277-285.SIMS, R. T. (1961). The blood vessels of the developing spinal cord of Xenopus laevis. J.

Embryol. exp. Morph. 9, 32-41.STOETER, P., SCHMIDT-LADEMANN, S. & VOIGT, K. (1978). Comparative microangiographic

and histologic studies of embryonal development of intracerebral capillaries. Neuro-radiology 16, 620-624.

STRONG, L. H. (1947). Primitive blood vessels of the spinal medulla of the rabbit, injectedwhile alive. Anat. Rec. 97, 58.

STRONG, L. H. (1961). The first appearance of vessels within the spinal cord of the mammal:their developing patterns as far as partial formation of the dorsal septum. Acta anat. 44,80-108.

STRONG, L. H. (1964). The early embryonic pattern of internal vascularization of the mam-malian cerebral cortex. /. comp. Neur. 123, 121-138.

STURROCK, R. R. (1981). A quantitative and morphological study of vascularization of thedeveloping mouse spinal cord. /. Anat. 132, 203-221.

{Accepted 10 March 1983)