Gradients of Histogenesis in the Ependymal Lining of the Third Ventricle in the Rabbit
VICTOR FERNANDEZ 1 and RODRIGO O. KULJIS 2
I Department of Physiology and Biophysics, Medical School, University of Chile, Santiago (Chile) and 2Section of Neuroanatomy and Department of Neurology, Yale University School of Medicine, New Haven, CT06510 (U.S.A.)
Tritiated thymidine autoradiography was used to analyze the site, time of origin, and developmental gradients of the specialized lin- ing of the ependymal surface of the third ventricle. Cells destined to form the ependyma are generated between days 15 and 22 of em- bryogenesis (gestation: 30+2 days), the majority of the cells undergoing final division on the 18th day of gestation. Ependymal cells originate in an orderly fashion according to 3 gradients. Two gradients of opposite direction (ventrodorsal and dorsoventral) are found in the parasaggital plane. Both gradients start at the level of the hypothalamic sulcus, progressively departing from this anatomical landmark as histogenesis progresses. A third gradient occurs in the caudorostral axis, such that cells located in caudal regions originate earlier than those located in rostral sectors. Thus, an orderly relationship exists between the time of origin of ependymal cells and their final location within the lining of the ventricular wall. These findings indicate, once again, the topographic nature of the gradients of histogenesis. The histogenic gradients displayed by the ependymal lining of the third ventricle appear strongly related to those exhi- bited by other diencephalic derivatives. The latter suggests that common factors govern the developmental sequence of all diencepha- lic derivatives as a function of their relative topographic location, independently of their functional role in the adult.
INTRODUCTION
The diencephalic ependyma has been the subject of much attention during the last 25 years, especially regarding the role of the subcommisural organ and
the hypothalamic ependyma in metabolic and endo- crine regulatory functions 6'11-13,15,17. Abundant in-
formation exists on the embryogenesis of various di-
encephalic derivatives since the advent of [3H]thymi- dine autoradiographic methods 1-4'16't8. However ,
comprehensive data on the correlation between the
histogenesis in thalamic and hypothalamic areas with the development of the adjacent ependymal lining
are still incomplete, to the best of our knowledge.
In the present study we analyzed the sequence of
histogenesis of the ependymal lining of the third ven- tricle by means of [3H]thymidine autoradiography.
Our observations indicate that this region of the dien- cephalon develops following orderly patterns of cell origin. The ependymal pattern of histogenesis ap- pears strongly related to the topographic location of
any given region of this epithelium within the dience-
phalon, such that it displays common sequences of
cell origin with physiologically diverse - - but adja- cent - - nuclear groups.
MATERIALS AND METHODS
New Zealand Black pregnant rabbits received a
single i.p. injection of 5 mCi/kg b. wt. of [3H]thymi-
dine (specific activity 6.0 Ci/mM, Schwarz Biore-
search) on days 15 (onset of cortical, basal telenceph-
alic nuclei and thalamic histogenesis) to 22 (end of
peak of cortical histogenesis) of gestation. At differ- ent survival periods 53 embryos were removed after
the injection of the tracer (Table I). Brains were
fixed by immersion, using 70-80% ethyl alcohol dur-
ing 1-2 days. Tissues were dehydrated, cleared in ce-
dar oil, and embedded in paraffin. Coronal sections 10 ktm thick were coated with NTB-3 Kodak nuclear emulsion and then exposed in the dark for 40 days at
4 °C. The coated sections were developed in dilute
Correspondence: R.O. Kuljis, Section of Neuroanatomy, Yale University School of Medicine, SHM C-303,333 Cedar Street, New Haven, CT 06150, U.S.A.
containing hardener . Af ter several rinses in distilled
water, the sections were stained with buffered fuch-
sin at pH 3.65, and mounted according to the proce-
dure of Kopriwa and Leblond 14. A summary of the
animals used and the age of the embryos at the time
of injection and sacrifice is shown in Table I.
The epithelial lining of the third ventricle was out-
lined at 4 selected, evenly spaced, coronal section
levels across the d iencephalon, using a camera luci-
da. Then, radioact ive cells with reduced silver grains
overlying the nucleus in densities significantly above
the background level were plot ted. Cells with 5 or
more evenly distr ibuted reduced silver grains overly-
ing the nucleus were arbi trar i ly considered "lightly
labeled ' . Cells with clumps of silver grains were des-
ignated 'heavily labeled ' and cells with enough grains
to prevent the individual recognit ion of each grain
'very heavily labeled ' . The lat ter arbi t rary conven-
tion permi t ted an object ive assessment of relative la-
beling intensity along the an te ropos te r ior and dorso-
ventral axes of the ependymal lining at different
stages of embryogenes is (Figs. 1-5) .
RESULTS
In rabbits exposed to [3H]thymidine on embryonic
days 15 ( E l 5 ) and 16 (E l6 ) , and examined la ter ac-
cording to Table I, labeled cells are localized pre-
dominant ly at caudal levels (Fig. 1, section 452; Fig,
2, section 448; Fig. 3A). Few, and only lightly labeled
cells are found in the frontal regions of the ependy-
mal epi thel ium (Fig. 1, sections 315 and 352; Fig. 2.
sections 301-350). The lat ter lightly labeled cells are
dis tr ibuted along the tha lamo-hypotha lamic border
TABLE I
Summary of animals used and the time intervals between injec- tion and sacrifice
Injection o f Number o f embryos on day o f removal
[3H]thymi- El5 El7 El8 El8 El9 E21 E23 E27 new- dine on days +3h +3h born
El5 2 1 - - 2 2 1 1 2 El6 + - 1 - 1 1 1 4 4 E 17 - - - t 1 1 3 6 E 18 " 1 t 1 5 9 E 20 . . . . . . . . 3 E 22 . . . . . . . 3
290
I 1 r a m I
' \ + /
315
.~+
t
(/,
352 4 5 2
Fig. 1. Drawings of serial coronal sections through the cpcndy- mal lining of the third ventricle showing the distribution of labeled cells. Tritiated thymidine was injected on El5. The em- bryo was sacrificed on E27. The lowest numbers identify the most frontal sections, and the highest numbers the most caudal section in the series. Radioactive cells are depicted around the outline of the ventricular cavity. Open circles, lightly labeled cells; filled circles, heavily labeled cells; triangles, very heavily labeled cells. See Materials and Methods for a description of in- tensity of labeling criteria.
at this period. More cells are labeled on E l 6 than on
the preceding day, and a larger number of them are
heavily labeled (Fig. 2). Heavily labeled cells are
present in animals injected during the lat ter stages in
the dorsa lmost region of the caudal ependyma (Fig.
1, section 452; Fig. 2, section 448). This area corre-
sponds to that of the subfornical organ in the adult.
In animals injected in E l 7 (Fig. 4), the amount of
I 1 m m j 4 4
350
Fig. 2. Drawings of coronal sections through the ependymal lin- ing of the third ventricle showing the distribution of labeled cells. Tritiated thymidine was injected on El6. The embryo was sacrificed on E27. Symbols as in Fig. 1
203
Fig. 3. Photomicrographs of autoradiograms of transverse sections through the E27 diencephalon at the level of the ependymal lining of the third ventricle. A: numerous lightly labeled cells are found in the caudal sector after injecting [3H]thymidine on El5. B: numer- ous heavily labeled cells are found in the frontoventral sector. C: A relatively sharp border between the labeled and unlabeled regions of the ependyma (arrow) occurs frontoventral to the hypothalamic sulcus. D: numerous heavily labeled cells are found in the region of the infundibulum. B-D correspond to cases that received [3H]thymidine at El8. The ependymal wall is parallel to the bottom edge of the panel in A-C, and runs obliquely from the upper left corner to the middle of the bottom in panel D. VC, ventricular cavity. Basic fuchsin counterstain. Bars: A, B and D = 25 ~tm, C = 30/~m.
o •o o ° ° ~ • o o • o • • • •
o o
.. ..o - o Oeo • •
2 9 3
"t 'i ! :
3 3 2
I 1 r a m I
o • o o e o o
V o % • • • o- .o ,I i
• o
. . ~ o" • "o o • • •
• ~o o o
e o Q • • ~
e
• o • • •
3 5 4 4 4 7
labeled cells increases significantly at frontal and me-
dial levels (sections 293,332 and 354). In the caudal
region (Fig. 4, section 447) and around the thalamo-
hypothalamic border (section 354), however, a large
proport ion of the posterior ependymal wall appears
unlabeled. The absence of labeled cells in this pre-
viously histogenically active area indicates that this
region has completed histogenesis by this stage.
Since large caudal sectors and the hypothalamic sul-
cus are the first to complete histogenesis, they proba-
Fig. 4. Drawings of coronal sections through the ependymal lin- ing of the third ventricle showing the distribution of labeled cells. Tritiated thymidine was injected in El7. The embryo was sacrificed in E27. Symbols as in Fig. 1.
204
bly represent the site from which the different histo- genic gradients of the ependymal epithelium origi- nate.
In animals injected on El8, the histogenic process
seems to accelerate in the diencephalon (Fig. 5).
Most cells present in the ependyma of the third ven-
tricle at this stage are very heavily labeled (Fig. 3B),
indicating that they are at or close to their final phase
of DNA synthesis at the time the [3H]thymidine was administered. In the frontal aspect of the third ventri-
cle wall, heavily labeled cells occupy the entire dor-
soventral extent of the ependymal lining, except for
the hypothalamic sulcus (Fig. 5, section 290). To-
wards the caudal region of the third ventricle, heavily
labeled cells are concentrated in progressively more
dorsal and ventral sectors of the ependyma (Fig. 3C,
D). A large intermediate region between the latter
sectors is free of radioactive material, and corre-
sponds to an area more advanced in histogenesis than
the frontoventral and frontodorsal ependyma (Fig. 5,
sections 329-455). The region of the subfornical or-
gan, that appears heavily labeled in animals injected
during E15-17, appears unlabeled in animals in-
jected after these stages (Fig. 6).
The litters of rabbits injected similarly in subse- quent stages of gestation show a very poor uptake of
[3H]thymidine into ependymal cells. Few labeled
cells lie in restricted areas in the dorsal and ventral
areas. Although autoradiography does not distin-
guish between neuronal and non-neuronal elements,
• =
i •
2 Q O • • ',.o
I 1 mWI • 3 2 9
, • e 3 5 4 4 5 5
Fig. 5. Drawings of coronal sections through the ependymal lin- ing of the third ventricle showing the distribution of labeled cells. Tritiated thymidine was injected in El8. The embryo was sacrificed on E27. Symbols as in Fig. 1.
- j
3 1 2 3 4 1
I l m m J
( ~ . . . . ¢ f \ / - - ) -"-
' J "i!
/ /
3 6 8 4 7 9
Fig. 6. Drawing of coronal sections through the ependymal lin- ing of the third ventricle showing the distribution of labeled cells. Tritiated thymidine was injected on E20. The subject was sacrificed at birth. Symbols as in Fig. 1.
some of the labeled cells clearly form part of vascular
walls, and thus do not seem to be of neural origin,
DISCUSSION
The cells destined to form the diencephalic epen-
dyma originate from the germinal matrix surround-
ing the third ventricle, following an orderly sequence
of histogenesis. The earlier stage of cell formation
takes place between El5 and El6, when cells located
in caudal regions in and around the thalamohypotha-
lamic border (hypothalamic sulcus) are formed. Mas- sive histogenesis takes place between El7 and El8,
when the cells destined to the frontal and middle sec- tors of the ependyma - - except the hypothalamic sulcus - - are formed. The large amount of heavily
labeled cells observed in the ependyma, in animals
injected in El8, indicates that the majority of the di-
encephalic ependymal cells are generated at this
time. A third stage takes place between E20 and E22, when histogenesis declines rapidly to finally cease almost completely. Only very few labeled cells are seen in the extreme dorsal and ventral regions of
the ependyma thereafter. These observations indicate the presence of 3 gra-
dients of histogenesis in the diencephalic ependyma. One gradient follows the caudorostral direction. starting at the caudal level of the hypothalamic sul- cus. Two other gradients in the parasaggital plane
205
also depart from the hypothalamic sulcus, one in the
ventrodorsal direction (thalamic ependyma), the
other in the dorsoventral direction (hypothalamic ependyma). Accordingly, the earliest cells to origi-
nate lie in and around the posterior aspect of the hy-
pothalamic sulcus. Cells originating later lie progres- sively more rostrally, dorsal or ventral to the hypo-
thalamic sulcus. The region of the subfornical organ
appears to constitute an exception to the topographic
gradients described. In fact, this region appears labeled in animals injected in the early, but not in the
advanced stages of histogenesis of the diencephalic ependyma. Contrary to cells destined to form nuclear
groups, diencephalic ependymal cells need not un-
dergo migratory movements to reach their definitive
location in the adult brain, and they remain at their
site of origin,
Except for the hypothalamic suicus and the region
of the subcommisural organ, the orientation of the
histogenic gradients displayed by the diencephalic
ependyma appear to be independent of anatomical
landmarks, and of the functional role that the various
diencephalic structures are to play in the adult. Thus, the time of origin of typical ependymal cells and of
some specialized lining cells appears largely subordi-
nated to their topographic location within the ven-
tricular wall, and totally independent of their functio-
nal roles. These histogenic characteristics are not pe-
culiar to the diencephalon. In fact, in telencephalic
development, for example, the rhinal fissure is the
sole anatomical landmark that seems to determine
the direction of histogenic gradients 9. Apart from the
rhinal fissure, telencephalic histogenesis takes place
independently of both anatomical landmarks and the adult functional roles of the structures being formed,
depending strictly only on the topographic location of
the developing structures in the telencephalic vesi-
cle 1°. Previous studies have shown that histogenic
gradients are also independent of the pattern of syn-
aptic connectivity displayed by a given cellular group in the adult 7.
The histogenic gradients displayed by the dien-
cephalic ependyma described above are closely re- lated with the gradients of cell deposit observed in di-
encephalic nuclear groups. In fact, the caudorostral, dorso-ventral and ventro-dorsal gradients described
here for ependymal cell formation correspond very closely with those previously reported in the thalamic
and hypothalamic nuclei 7's. As could be expected
from previous studies on the development of other
areas of the prosencephalon, these findings indicate
that histogenic gradients in the diencephalon are
strictly determined by the topographic location of the
various cell groups within the vesicle, and are largely
independent of the future functional role of the areas under formation.
ACKNOWLEDGMENTS
This study was supported by Proyectos B-
1583/8215, B-1538/8325, B-1538/8435 and B-
1538/8535, University of Chile.
REFERENCES
1 Altman, J. and Bayer, S.A., Development of the dience- phalon in the rat. I. Autoradiographic study of the time of origin and settling patterns of neurons of the hypothalamus, J. Comp. Neurol., 182 (1978) 945-972.
2 Altman, J. and Bayer, S.A., Development of the dience- phalon in the rat. II. Correlation of the embryonic devel- opment of the hypothalamus with the time of origin of its neurons, J. Cornp. Neurol., 182 (1978) 973-994.
3 Altman, J. and Bayer, S.A., Development of the dience- phalon in the rat. III. Ontogeny of the specialized ventricu- lar lining of the hypothalamic third ventricle, J. Comp. Neurol., 182 (1978) 995-1016.
4 Angevine, J.B., Jr., Time of neuron origin in the dience- phalon of the mouse. An autoradiographic study, J. Comp. Neurol., 139 (1970) 129-188.
6 Bleier, R., Surface fine structure of subependymal cells of
the hypothalamic third ventricle of the mouse, J. Cornp. Neurol., 161 (1975) 555-568.
7 Fern~indez, V., An autoradiographic study of the devel- opment of the anterior thalamic groups and limbic cortex in the rabbit, J. Comp. Neurol., 136 (1969) 423-452.
8 Fern~indez, V. and Bravo, H., Autoradiographic study of the development of the medial thalamic nuclei in the rabbit, Brain, Behav. Evol., 7 (1973) 453-465.
9 Fern~indez, V. and Bravo, H., Autoradiographic study of the development of the cerebral cortex in the rabbit, Brain Behav. Evol., 9 (1974) 317-332.
10 Fern~indez, V., Bravo, H., Kuljis, R.O. and Fuentes, I., Autoradiographic study of the development of the neostria- tum in the rabbit, Brain Behav. Evol., 16 (1979) 113-128.
11 Ives, P.J. and McArthur, N.H., The armadillo infundibu- lum: correlative histochemistry, scanning and transmission electron microscopy of the ventricular surface, Brain Res., 190 (1980) 385-397.
12 Isom~iki, A., Kivalo, E. and Talanti, S., Electron micro- scopic structure of the subcommisural organ in the calf (Bos
206
taurus) with special reference to secretory phenomena, Ants. Acad. Scient. Fenn., 111 (1965) 1-64.
13 Kimble, J.E. and Mollg/ird, E., Evidence for basal secre- tion in the subcommisural organ of the rabbit. Z. Zell- ]brsch., 142 (1973) 223-239.
14 Kopriwa, B.M. and Leblond, C.P.. Improvements in the coating technique for radioautography, J. Histochem. Crto- chem., 10 (1963) 269-284.
15 L6fgren, F., The infundibular recess, a component of the hypothalamo-adenohypophyseal system. Acta Morph.
Neer/-S'copu/, 3(Jc459)53 2">;. [6 Rakic, P. and ,~ id l l lan , R [ . , S t ibcol l ln l i sur~l [ ~l }~,lll ~ltci t , , f
jacent ependyma autoradiographic ~,ltlti\ Ol ! b o i l ~,lieill ilt
the mouse brain..tin..I..t11~ll., t 22 ( [96S ) 317 a ~6 17 Rhoda. 1., Corbicr, P. and Roffi, .I,. t l},potha];mm *cst~s
teronc increase m male rat at birth, l~it ,1 De; \~',r~s, i.. i ( 19831 187-190,
18 Shimada, M. and Nakamura, I . . H m c of ncur, m orimn m mouse hypothalamic nuclei, Exp. Neuro/ 1! (1~)73) 163- 173.
