Acta l~eurochirurgica 31, 153--159 (i975) 9 b y Springer-Verlag 1975
I. Anatomisches I n s t i t u t der Universit/~t des Saarlandes
(Direktor: Prof. Dr. H. Leonhardt), Homburg (Saar)
Recent Observations on Ependyma and Subependymal Basement Membranes* ** By
H. Leonhardt and U. Desaga W i t h 3 Figures
Summary Directly under the ependyma, and between the ependymal cells, there are basement membranes which form labyrinths connecting with the perivascular basement membranes of the subependymal vessels. The basement membranes exhibit differences in form, position, and distribution, and t h e y can be distended b y fluid absorption into large lacunae. The membranes can be identified as glyeoprotein and glyleolipid substances. Electron microscopic studies have shown the differences between the basement membrane labyrinths of the human, the r a b b i t and the rat. I n the human, the labyrinths contain isolated collagen fibrils. Basement membranes generally line all connective tissue spaces and form their interstitial borders. F r o m this point of view, the subependymal basement membrane labyrinths can be seen as interstitial spaces near the ventricles, forming pathways between the ependyma and the subependymal capillaries or postcapillary veins. One can p r o c e e d from t h e a s s u m p t i o n that t h e g r e a t e r p o r t i o n of t h e cerebrospinal fluid is p r o d u c e d in t h e ventricles, while t h e lesser p o r t i o n is f o r m e d e x t r a v e n t r i c n l a r l y . According to Bering (1965), 55~o of t h e cerebrospinal fluid in children is p r o d u c e d in t h e ventricles, while 45 ~o originates e x t r a v e n t r i c u l a r l y . These p e r c e n t u a l r e l a t i o n s h i p s are similar to those f o u n d in t h e dog, t h e cat, a n d t h e r a b b i t (Bering a n d S a t e 1963, P o l l a y a n d D a v s o n 1963). Since t h e t i m e s of Cushing (1914) * This investigation was supported b y a grant from the "Deutsche Forsehungsgemeinsehaft". ** The paper has been presented at the Meeting of the Society for Research into Hydrocephalus and Spina Bifida, Rotterdam, 28th June to 1st J u l y 1972. Acta Neurochirurgica, Vol. 31, Fasc. 3--4
11
154
H. Leonhardt and U. Desaga:
and Dandy (1919) it has been generally accepted that the intraventricular cerebrospinal fluid is produced by the choroid plexuses (de Rougemont, Ames, Nesbett, and Hoffman 1960, Welch 1963). There are also good reasons for believing that a large quantity of cerebrospinal fluid enters the ventricles through the ependyma from the ventricular walls (Jaeobi and Magnus 1925, Bering and Sate 1963, Bering 1965). On the basis of extensive computations Bering and Sate (1963) arrived at the conclusion that approximately half of the intraventricularly produced cerebrospinal fluid penetrates the ventricular wall. A detailed description of the problem has been presented by Bering (1965). With regard to the problems of aetiology and therapy of hydrocephalus, the movement of fluid between the ventricles and subependymM blood vessels is of particular interest. Brightman (1965) has shown that in the rat ferritin enters the ependyma from the ventrieular fluid through micropinocytosis and is subsequently transported through these cells. At the basal ends of the ependymal cells, the ferritin is concentrated in "dense filler". From here, it enters the blood vessels. In our experiments we examined this pathway from the basal ends of the ependymai cells to the blood vessels in the rabbit, the rat, and recently in the human being (Leonhardt 1970, 1972, Desaga 1971, 1972), and were able to make the following observations; directly under the ependyma as well as between the ependymal cells the basement membrane forms labyrinths in a number of situations. Furthermore, the basement membranes have connections with the perivascular basement membranes of the subependymal capillaries and postcapillary veins. Using the new periodic acid bisulfit aldehydthionine staining method (Specht 1970), a light microscopic differentiation between the blue-stained basement membranes and the red-stained brain parenchyma is possible. The basement membranes can, utilizing this technique, be simultaneously identified as glycoprotein and gtyeolipid substances, which through swelling can hold fluid. The basement membranes exhibit differences in form, position, and distribution. They may be short and star-shaped, or long, drawn--out structures lying either under the ependyma or near the ependymal surface. Their topographic distribution shows a preponderance in the fissure-like angles and narrow portions of the ventricular system where one could normally expect vortex formation and a reduction in the flow speed (Fig. 1). In the rat, numerous basement membranes are observed in the caudal end of the central canal (Booz and Desaga 1972). Electron microscopic studies have shown the differences between the basement membrane labyrinths of the rabbit and the rat. The basement membrane of the rabbit is characterized by a lamina densa and two laminae rarae, while that of the rat has but one electron dense
1R,ecent Observations on Ependsana
155
Figs. i a-c. Sub- and interependymal basement membrane labyrinths in rabbit, a Subependymal labyrinths (arrows), 3ra ventricle, X 150. b Interependymal labyrinths (arrows), lateral ventricle, • 300. c Subependymal labyrinths (l) in connection with the perivaseular basement membrane (pbm), lateral ventricle, • 1,200. d Position of interependymal labyrinths between the ependymal cells, cross-section, lateral ventricle, • 1,200. PBA-staining (Speeht)
11"
Figs. 2 a-c. S u b e p e n d y r n a l b a s e m e n t m e m b r a n e l a b y r i n t h s in rabbit, lateral ventricle, a N a r r o w a n d b r a n c h e d labyl"inths (1) below the e p e n d y m M surface (4), • 18,000. b L a b y r i n t h (1) distended as a large l a c u n a below the e p e n d y m a l surface (e). c Subependyrnal l a b y r i n t h (l) in connection w i t h t h e perivascul~r b a s e m e n t m e m b r a n e (pbm) of a capillary (c), • 18,000. 2 d-e. S u b e p e n d y r n a l b a s e m e n t m e m b r a n e l a b y r i n t h s in rat, 3rd ventricle. d N a r r o w a n d b r a n c h e d l a b y r i n t h (1), • 19,000. e D i s t e n d e d l a b y r i n t h (1), • 19,000
Figs. 3 a - b . S u b e p e n d y m a l b a s e m e n t m e m b r a n e l a b y r i n t h s in h u m a n , 4~TM ventricle, a Narrow a n d branches l a b y r i n t h (l) with eollagenous fibers if), • 27,000. b D i s t e n d e d l a b y r i n t h (1) with eollagenous fibers (f), • 54,000
158
H. Lconhardt and U. Desaga:
mass which completely fills the intercellular space (see also Westergaard /970). The basemen~ membranes can, however, be distended as large lacunae through fluid absorption (Fig. 2). Thus, the narrow branched labyrinths can be seen as a potential fluid storage space which can be filled with ventricular fluid through membrane vesieulation of the ependymal cells as well as through the connections between the perivascular basement membrane with the capillary wall. I n the meantime, we have been able to expand our investigation to the h u m a n ventricular system. Subependymal basement membranes in varying distribution and density are found here also. I n humans, however, the basement membrane also forms a p a t h w a y between ependyma and subependymal capillaries or postcapillary veins (Fig. 3). In this respect, the relationships between the human, the rat, and the rabbit are comparable. The fine structure, however, exhibits a marked difference between the h u m a n basement membrane labyrinth and t h a t of the rat in t h a t it has a stratified basement membrane. A lamina rara externa, lamina densa, and a lamina rara interna are recognizable. The main difference between the labyrinths of man and rabbit is t h a t h u m a n labyrinths contain isolated collagen fibrils. W h a t further studies of function are suggested by these observations .~ Basement membranes are generally present in regions where material interchange occurs (see Vollrath 1968). For example, they are found at the basal ends of surface epithelia in structures such as the intestine. The basement membrane is also an integral part of the kidney filter and covers muscle cells as well as lining capillaries. Furthermore, basement membranes line connective tissue spaces and form interstitial borders. From this point of view, the basement membrane labyrinths can be seen as interstitial spaces in the neighbourhood of ventricles. Could fluid transport through these spaces be enzymatically increased or inhibited as it is in other areas by the action of hyaluronidase? Specific experiments in this direction have been started.
Acknowledgement The authors thank Professor Loew (Hornburg/Saar) for the operation material. We also gratefully acknowledge valuable technical assistance from Miss Elisabeth Oestermann and Mrs. ielga Zuther-Witzsch.
Recent Observations on E p e n d y m a
159
References Bering, E . A . Jr., The eerebrospinal fluid circulation. I n : Cerebrospinal fluid and the regulation of ventilation. The proceedings of a symposium held at the Downstate Medical Center, State University of New York (Ch. MeC. Brooks, F. F. Kao, and B. B. Lloyd ed.). Oxford: Blaekwell Scientific Publications. 1965. - - and O. Sate, Hydrocephalus: changes in formation and absorption of eerebrospinal fluid within the cerebral ventricles. J. Neurosurg. 20 (1963), 1050--1063. Booz, K. It., and U. Desaga, Sub- u n d interependymale Basalmembranlabyrinthe am Ventrikelsystem u n d ZentralkanM der wei2en Ratte. Anat. Anz. 135, Erg.-Bd. (1973), 609---612. Brightman, 3/[. W., The distribution within the brain of ferritin injected into eerebrospinal fluid compartments. I. Ependymal distribution. J. Cell. Biol. 26 (1965), 99--123. Cushing, H., Studies on the cerebrospinal fluid. J. med. Bes. 31 (1914), 1--49. Band?-, W. E., Where is cerebrospinM fluid absorbed ? J. Amer. med. Ass. 92 (1929), 2012--2014. De gougemont, J., A. Ames I I I , F. B. Nesbett, and H. F. Hofmann, Fluid formed by ehoroid plexus. J. Neurophysiol. 23 (1960), 485--495. Desaga, U., Lieht- u n d elektronemnikroskopiseher Naehweis subependymaler Basalmembranlabyrinthe im I I I . Ventrikel der Ratte. Z. mikr.anat. Forseh. 85 (1972), 50--54. - - F o r m u n d Verteilung subependymMer Basalmembranlabyrinthe am Ventrikelsystem der Ratte. Z. Zellforseh. 132 (1972), 553--562. Jacobi, W., and G. Magnus, Gefgl3 u n d Liquorstudien am Hirn des lebenden Hundes. Arch. Psychiat. Nerv. Krankh. 73 (1925), 126--138. Leonhardt, It., Subependymale Basalmembranlabyrinthe im ttinterhorn des Seitenventrikels des Kaninehengehirns. Zur Frage des Liquorabflusses. Z. Zellforseh. 105 (1970), 595--604. - - Ober die topographisehe Verteilung der subependymalen Basalmembranlabyrinthe im Ventrikelsystem des Kaninehengehirns. Z. Zellforseh. 127 (1972), 392--406. - - Uber elektronenmikroskopisehe Untersehiede zwischen don subependymMen BasMmembranlabyrinthen yon Mensch, Ratte und Kaninehen. Anat. Ariz. 135, Erg.-Bd. (1973), 605--607. Pollay, 3/[., and H. Davson, The passage of certain substances out of the eerebrospinal fluid. Brain 86 (1963), 137--150. Speeht, W., Fgrben mit Aldehydthionin. Eine Methode fiir den topoehemisehen Naehweis yon Sulfons~uren u n d Aldehyden. 65. Vers. Anat. Ges., Wiirzburg 1970. Vollrath, L., Uber Bau u n d F u n k t i o n yon BasMmembranen. Dtseh. reed. Wsehr. 93 (1968), 360--365. Welch, 2 . , Secretion of eerebrospinal fluid b y ehoroid plexus of the rabbit. Amer. J. Physiol. 205 (1963), 617--624. Westergaard, E., The lateral ventricles and the ventrieular walls. An anatomical, histological and electron-microscopic investigation on mice, rats, hamsters, guinea-pigs and rabbits. I kommission hos Andelsbogtrykkeriet i Odense, 1970. Authors' address: Prof. Dr. H. Leonhardt, Anatomisehes I n s t i t u t der Universit/~t, Olshausenstrage, D-2300 Kiel, Federal l~epublie of Germany.
I. Anatomisches I n s t i t u t der Universit/~t des Saarlandes
(Direktor: Prof. Dr. H. Leonhardt), Homburg (Saar)
Recent Observations on Ependyma and Subependymal Basement Membranes* ** By
H. Leonhardt and U. Desaga W i t h 3 Figures
Summary Directly under the ependyma, and between the ependymal cells, there are basement membranes which form labyrinths connecting with the perivascular basement membranes of the subependymal vessels. The basement membranes exhibit differences in form, position, and distribution, and t h e y can be distended b y fluid absorption into large lacunae. The membranes can be identified as glyeoprotein and glyleolipid substances. Electron microscopic studies have shown the differences between the basement membrane labyrinths of the human, the r a b b i t and the rat. I n the human, the labyrinths contain isolated collagen fibrils. Basement membranes generally line all connective tissue spaces and form their interstitial borders. F r o m this point of view, the subependymal basement membrane labyrinths can be seen as interstitial spaces near the ventricles, forming pathways between the ependyma and the subependymal capillaries or postcapillary veins. One can p r o c e e d from t h e a s s u m p t i o n that t h e g r e a t e r p o r t i o n of t h e cerebrospinal fluid is p r o d u c e d in t h e ventricles, while t h e lesser p o r t i o n is f o r m e d e x t r a v e n t r i c n l a r l y . According to Bering (1965), 55~o of t h e cerebrospinal fluid in children is p r o d u c e d in t h e ventricles, while 45 ~o originates e x t r a v e n t r i c u l a r l y . These p e r c e n t u a l r e l a t i o n s h i p s are similar to those f o u n d in t h e dog, t h e cat, a n d t h e r a b b i t (Bering a n d S a t e 1963, P o l l a y a n d D a v s o n 1963). Since t h e t i m e s of Cushing (1914) * This investigation was supported b y a grant from the "Deutsche Forsehungsgemeinsehaft". ** The paper has been presented at the Meeting of the Society for Research into Hydrocephalus and Spina Bifida, Rotterdam, 28th June to 1st J u l y 1972. Acta Neurochirurgica, Vol. 31, Fasc. 3--4
11
154
H. Leonhardt and U. Desaga:
and Dandy (1919) it has been generally accepted that the intraventricular cerebrospinal fluid is produced by the choroid plexuses (de Rougemont, Ames, Nesbett, and Hoffman 1960, Welch 1963). There are also good reasons for believing that a large quantity of cerebrospinal fluid enters the ventricles through the ependyma from the ventricular walls (Jaeobi and Magnus 1925, Bering and Sate 1963, Bering 1965). On the basis of extensive computations Bering and Sate (1963) arrived at the conclusion that approximately half of the intraventricularly produced cerebrospinal fluid penetrates the ventricular wall. A detailed description of the problem has been presented by Bering (1965). With regard to the problems of aetiology and therapy of hydrocephalus, the movement of fluid between the ventricles and subependymM blood vessels is of particular interest. Brightman (1965) has shown that in the rat ferritin enters the ependyma from the ventrieular fluid through micropinocytosis and is subsequently transported through these cells. At the basal ends of the ependymal cells, the ferritin is concentrated in "dense filler". From here, it enters the blood vessels. In our experiments we examined this pathway from the basal ends of the ependymai cells to the blood vessels in the rabbit, the rat, and recently in the human being (Leonhardt 1970, 1972, Desaga 1971, 1972), and were able to make the following observations; directly under the ependyma as well as between the ependymal cells the basement membrane forms labyrinths in a number of situations. Furthermore, the basement membranes have connections with the perivascular basement membranes of the subependymal capillaries and postcapillary veins. Using the new periodic acid bisulfit aldehydthionine staining method (Specht 1970), a light microscopic differentiation between the blue-stained basement membranes and the red-stained brain parenchyma is possible. The basement membranes can, utilizing this technique, be simultaneously identified as glycoprotein and gtyeolipid substances, which through swelling can hold fluid. The basement membranes exhibit differences in form, position, and distribution. They may be short and star-shaped, or long, drawn--out structures lying either under the ependyma or near the ependymal surface. Their topographic distribution shows a preponderance in the fissure-like angles and narrow portions of the ventricular system where one could normally expect vortex formation and a reduction in the flow speed (Fig. 1). In the rat, numerous basement membranes are observed in the caudal end of the central canal (Booz and Desaga 1972). Electron microscopic studies have shown the differences between the basement membrane labyrinths of the rabbit and the rat. The basement membrane of the rabbit is characterized by a lamina densa and two laminae rarae, while that of the rat has but one electron dense
1R,ecent Observations on Ependsana
155
Figs. i a-c. Sub- and interependymal basement membrane labyrinths in rabbit, a Subependymal labyrinths (arrows), 3ra ventricle, X 150. b Interependymal labyrinths (arrows), lateral ventricle, • 300. c Subependymal labyrinths (l) in connection with the perivaseular basement membrane (pbm), lateral ventricle, • 1,200. d Position of interependymal labyrinths between the ependymal cells, cross-section, lateral ventricle, • 1,200. PBA-staining (Speeht)
11"
Figs. 2 a-c. S u b e p e n d y r n a l b a s e m e n t m e m b r a n e l a b y r i n t h s in rabbit, lateral ventricle, a N a r r o w a n d b r a n c h e d labyl"inths (1) below the e p e n d y m M surface (4), • 18,000. b L a b y r i n t h (1) distended as a large l a c u n a below the e p e n d y m a l surface (e). c Subependyrnal l a b y r i n t h (l) in connection w i t h t h e perivascul~r b a s e m e n t m e m b r a n e (pbm) of a capillary (c), • 18,000. 2 d-e. S u b e p e n d y r n a l b a s e m e n t m e m b r a n e l a b y r i n t h s in rat, 3rd ventricle. d N a r r o w a n d b r a n c h e d l a b y r i n t h (1), • 19,000. e D i s t e n d e d l a b y r i n t h (1), • 19,000
Figs. 3 a - b . S u b e p e n d y m a l b a s e m e n t m e m b r a n e l a b y r i n t h s in h u m a n , 4~TM ventricle, a Narrow a n d branches l a b y r i n t h (l) with eollagenous fibers if), • 27,000. b D i s t e n d e d l a b y r i n t h (1) with eollagenous fibers (f), • 54,000
158
H. Lconhardt and U. Desaga:
mass which completely fills the intercellular space (see also Westergaard /970). The basemen~ membranes can, however, be distended as large lacunae through fluid absorption (Fig. 2). Thus, the narrow branched labyrinths can be seen as a potential fluid storage space which can be filled with ventricular fluid through membrane vesieulation of the ependymal cells as well as through the connections between the perivascular basement membrane with the capillary wall. I n the meantime, we have been able to expand our investigation to the h u m a n ventricular system. Subependymal basement membranes in varying distribution and density are found here also. I n humans, however, the basement membrane also forms a p a t h w a y between ependyma and subependymal capillaries or postcapillary veins (Fig. 3). In this respect, the relationships between the human, the rat, and the rabbit are comparable. The fine structure, however, exhibits a marked difference between the h u m a n basement membrane labyrinth and t h a t of the rat in t h a t it has a stratified basement membrane. A lamina rara externa, lamina densa, and a lamina rara interna are recognizable. The main difference between the labyrinths of man and rabbit is t h a t h u m a n labyrinths contain isolated collagen fibrils. W h a t further studies of function are suggested by these observations .~ Basement membranes are generally present in regions where material interchange occurs (see Vollrath 1968). For example, they are found at the basal ends of surface epithelia in structures such as the intestine. The basement membrane is also an integral part of the kidney filter and covers muscle cells as well as lining capillaries. Furthermore, basement membranes line connective tissue spaces and form interstitial borders. From this point of view, the basement membrane labyrinths can be seen as interstitial spaces in the neighbourhood of ventricles. Could fluid transport through these spaces be enzymatically increased or inhibited as it is in other areas by the action of hyaluronidase? Specific experiments in this direction have been started.
Acknowledgement The authors thank Professor Loew (Hornburg/Saar) for the operation material. We also gratefully acknowledge valuable technical assistance from Miss Elisabeth Oestermann and Mrs. ielga Zuther-Witzsch.
Recent Observations on E p e n d y m a
159
References Bering, E . A . Jr., The eerebrospinal fluid circulation. I n : Cerebrospinal fluid and the regulation of ventilation. The proceedings of a symposium held at the Downstate Medical Center, State University of New York (Ch. MeC. Brooks, F. F. Kao, and B. B. Lloyd ed.). Oxford: Blaekwell Scientific Publications. 1965. - - and O. Sate, Hydrocephalus: changes in formation and absorption of eerebrospinal fluid within the cerebral ventricles. J. Neurosurg. 20 (1963), 1050--1063. Booz, K. It., and U. Desaga, Sub- u n d interependymale Basalmembranlabyrinthe am Ventrikelsystem u n d ZentralkanM der wei2en Ratte. Anat. Anz. 135, Erg.-Bd. (1973), 609---612. Brightman, 3/[. W., The distribution within the brain of ferritin injected into eerebrospinal fluid compartments. I. Ependymal distribution. J. Cell. Biol. 26 (1965), 99--123. Cushing, H., Studies on the cerebrospinal fluid. J. med. Bes. 31 (1914), 1--49. Band?-, W. E., Where is cerebrospinM fluid absorbed ? J. Amer. med. Ass. 92 (1929), 2012--2014. De gougemont, J., A. Ames I I I , F. B. Nesbett, and H. F. Hofmann, Fluid formed by ehoroid plexus. J. Neurophysiol. 23 (1960), 485--495. Desaga, U., Lieht- u n d elektronemnikroskopiseher Naehweis subependymaler Basalmembranlabyrinthe im I I I . Ventrikel der Ratte. Z. mikr.anat. Forseh. 85 (1972), 50--54. - - F o r m u n d Verteilung subependymMer Basalmembranlabyrinthe am Ventrikelsystem der Ratte. Z. Zellforseh. 132 (1972), 553--562. Jacobi, W., and G. Magnus, Gefgl3 u n d Liquorstudien am Hirn des lebenden Hundes. Arch. Psychiat. Nerv. Krankh. 73 (1925), 126--138. Leonhardt, It., Subependymale Basalmembranlabyrinthe im ttinterhorn des Seitenventrikels des Kaninehengehirns. Zur Frage des Liquorabflusses. Z. Zellforseh. 105 (1970), 595--604. - - Ober die topographisehe Verteilung der subependymalen Basalmembranlabyrinthe im Ventrikelsystem des Kaninehengehirns. Z. Zellforseh. 127 (1972), 392--406. - - Uber elektronenmikroskopisehe Untersehiede zwischen don subependymMen BasMmembranlabyrinthen yon Mensch, Ratte und Kaninehen. Anat. Ariz. 135, Erg.-Bd. (1973), 605--607. Pollay, 3/[., and H. Davson, The passage of certain substances out of the eerebrospinal fluid. Brain 86 (1963), 137--150. Speeht, W., Fgrben mit Aldehydthionin. Eine Methode fiir den topoehemisehen Naehweis yon Sulfons~uren u n d Aldehyden. 65. Vers. Anat. Ges., Wiirzburg 1970. Vollrath, L., Uber Bau u n d F u n k t i o n yon BasMmembranen. Dtseh. reed. Wsehr. 93 (1968), 360--365. Welch, 2 . , Secretion of eerebrospinal fluid b y ehoroid plexus of the rabbit. Amer. J. Physiol. 205 (1963), 617--624. Westergaard, E., The lateral ventricles and the ventrieular walls. An anatomical, histological and electron-microscopic investigation on mice, rats, hamsters, guinea-pigs and rabbits. I kommission hos Andelsbogtrykkeriet i Odense, 1970. Authors' address: Prof. Dr. H. Leonhardt, Anatomisehes I n s t i t u t der Universit/~t, Olshausenstrage, D-2300 Kiel, Federal l~epublie of Germany.
