Early Vs Delayed Ventriculoperitoneal Shunt-Effects on the Impairment of the Developing Brain in Congenitally Hydrocephalic HTX-Rats

Early Vs Delayed Ventriculoperitoneal Shunt-Effects on the Impairment of the Developing Brain in Congenitally Hydrocephalic HTX-Rats

Kikuo Suda, Kiyoshi Sato, Nobuaki Takeda, Mitsuru Wada, Takahito Miyazawa, Hajime Arai, Masanori Ito, and Makoto Miyaoka

Summary. In the present investigation, we report on the effects of the early placement of a ventriculoperitoneal shunt (V-P shunt) on the development of cerebral synapses by counting spine density of the cortical pyramidal neurons (stained by rapid Golgi method) and measuring one of the synaptic vesicle proteins, SVP-38. The techniques used were quantitative histochemical and immunoblot analysis. The learning ability of congenitally hydrocephalic HTX­ rats whose hydrocephalus had been arrested by insertion of a V-P shunt 7-9 days after birth (Early Shunt) was assessed by the light-darkness discrimina­ tion test. When a V-P shunt was inserted into the hydrocephalic animals approximately 4 weeks after birth (Delayed Shunt), not only was there no reduction in the size of the abnormally enlarged ventricles, but also there was no increase in cortical mantle thickness. Furthermore, spine density in the cerebral cortex in such animals was found to be decreased. Learning disability could not be corrected by the delayed shunt procedure. Contrary to these observations, early shunt placement was found to result in normalization of the abnormally enlarged ventricles, concomitant with simultaneous cortical mantle thickening and prevention of both decreased spine density and decay of SVP- 38 in the affected cerebral cortex. The learning disability of such animals was not found to be disturbed, compared with that of the sham-operation group. From these observations, it is concluded that early shunt placement may have a beneficial role not only in repairing, but also in preventing the impairment of synaptogenesis caused by the progression of congenital hydrocephalus.

Keywords. Congenital hydrocephalus - Rat - Ventriculoperitoneal shunt synaptogenesis -Learning ability

Introduction

In previous experiments using congenitally hydrocephalic HTX-rats (Kohn et a!. 1981, 1984), we reported that congenital hydrocephalus impaired the development of dendrites and spines of neurons in the affected cerebral cortex (Miyazawa et a!. 1988). Such impairment of neuronal development in the hydrocephalic brain of HTX-rats may not be completely corrected by insertion of a ventriculoperitoneal shunt approximately 4 weeks after birth (Delayed Shunt). We also suggested that the learning disability found in mature HTX­ rats whose hydrocephalus had been arrested by delayed shunt insertion could be related to the aforementioned disturbance of neuronal development, especially synaptogenesis of the brain (Miyazawa and Sato 1991).

In the present study, we report on the beneficial effects of early placement of a ventriculoperitoneal shunt (Early Shunt) in HTX-rats whose hydrocephalus was arrested by insertion of a ventriculoperitoneal shunt 7-9 days after birth. The beneficial effects on the development of cerebral synapses, and also on the learning ability of these animals, were examined.

Materials and Methods

Congenitally hydrocephalic male HTX-rats (HTX) were used, and non­ hydrocephalic male HTX-rats served as controls with and without sham operations.

Preparation of HTX with Early and Delayed Shunts

HTX manifesting hydrocephalus were divided into two groups in accordance with the time of ventriculoperitoneal (V-P) shunt insertion, that is, the Early Shunt Group in which a V-P shunt was intially inserted 7-9 days after birth, and the Delayed Shunt Group in which the V-P shunt was inserted approxi­ mately 4 weeks after birth. Non-hydrocephalic HTX were also divided into two groups, in one of which sham operations were carried out at the same times as the shunt operations in the experimental groups.

The HTX were anesthetized in the prone position by inhalation of 1.0% halothane. The scalp was incised to expose the parietal bone on the left side, and a hole approximately 2 mm in diameter was bored in the skull in an area 2 mm anterior to the left lambdoid suture and 4-5 mm to the left of the sagittal suture. After the dura matter was exposed and coagulated, a laparatomy was performed at the right dorsal flank to expose the peritoneal space. A shunt passer was inserted into the subcutaneous tissue, and then a V-P shunt tube without pressure regulation valve (Dow Corning Co., silastic catheter, inside diameter 0.025 in, length 12em) was passed from the left parietal area to the right flank. The tip of the ventricular catheter was inserted into the left lateral ventricle 4-5 mm from the inner table of the calvaria. After the flow of the cerebrospinal fluid (CSF) from the abdominal catheter was confirmed, the ventricular catheter was fixed to the skull using Aron Alpha. The abdominal catheter was then inserted into the peritoneal space, and the skin incision was closed.

Ventriculography and Magnetic Resonance Imaging (MRI)

The chronological change in ventricular size and thickness of the cortical mantle of the brains of animals subjected to a V-P shunt was assessed at random times after the surgery by ventriculography and by MRI (installed at the National Institute for Physiological Science in Okazaki). During ven­ triculography under Halothane general anesthesia, 0.2-0.3 ml of Iotroran was slowly injected into the lateral ventricle via V-P shunt abdominal catheter or via 27 G scalp needle inserted directly into the lateral ventricle. Softex X-ray films (Softex Corp., Tokyo) were used to obtain ventriculograms, and the lateral ventricular size visualized on the Softex film was measured by planimetry using a computer-assisted image analyzer (!BASS 2000, Zeiss). The MRI used in this study was Hitachi 2.114 T, and T1 weighted images were obtained.

Light-Darkness Discrimination using Y-Maze Test (Takiguchi et al. 1988)

When the experimental and sham-operated animals subjected either to an early or a delayed V-P shunt became sexually mature, their learning ability was assessed by the Y-maze test, as described elsewhere (Miyazawa and Sato 1991). A mean correct response rate and a mean response latency time were calculated for each set of 10 trials for each respective animal. Different condi­ tions in the animals were compared statistically, using three-way analysis of variance (ANOVA) and the Wilcoxon t-test.

Measurement of Locomotor Activity

All animals subjected to the Y-maze test were individually placed on an Automex-II in order to count their motions in a 12-hour period from 19:00 to 07:00. The statistical differences in motion counts among the different experi­ mental animal groups were evaluated using the Wilcoxon t-test.

Neuropathological Study

Animals which had completed the Y-maze test and assessment of locomotor activity were anesthetized by ethyl inhalation. They were sacrificed by trans­ cardial perfusion with a 4% paraformaldehyde solution for conventional light microscope and histochemical examinations, and with a perfusate containing 3% potassium dichromate and 0.2% osmium tetroxide for quantitative Golgi study (Millhouse 1981). In the latter study, the spines of the apical and basal dendrites of the pyramidal neurons in layers II and III of the fronto-parietal cortex (Zilles and Wree 1985) were quantitatively measured, as described elsewhere (Sholl1953; Miyazawa et al. 1988). In brief, the number of spines on 20 J.tm segments of, apical and basal dendrites respectively, at a distance of 100 J.tm from the cell body of a neuron were counted. The number of dendritic spines obtained from 30 pyramidal neurons of one aminal was considered to represent spine density. Student's t-test was used for the statistical comparison of spine densities among different experimental groups (Miyazawa and Sato 1991).

Paraffin-embedded sections of the brain were stained by an indirect immuno­ histochemical technique, with mouse monoclonal antibody (MAb) against synaptic vesicle protein (SVP-38) purified from guinea-pig cerebrum (Obata et al. 1986, 1987); histochemical quantification of SVP-38 was then carried out. In brief, the sections were incubated with MAb and FITC-conjugated sheep anti-mouse IgG, after which the fluorescence intensity of the immunoreactive products in a 5-J.tm spot in the molecular layer of the parieto-occipital cortex was measured with a microphotometer. The fluorescence intensity values were expressed as the mean standard error of the 60 measurements of four rats in each experimental group, and the results were statistically compared by Student's t-test (T. Miyazawa et al., unpublished data).

Immunoblot Analysis of Synaptic Vesicle Protein (SVP-38)

For electrophoresis, animals in each experimental group were sacrificed by decapitation at 4 weeks of age. The isolated brain tissues were homogenized in 10 volumes of electrophoresis-sample buffer and subjected to immunoblot analysis. Sodium dodecy1 sulfate polyacrylamide gel electrophoresis (SDS­ PAGE) was performed by Laemmli's method (1970). The gels were then electroblotted onto nitrocellulose sheets and the blots were reacted successively with MAb, HRP-conjugated goat anti-mouse IgG, and 3,3'-diaminobenzidine (DAB).

Results

Sequelae of Early Ventriculoperitoneal Shunt

The survival rates of 4-week and 8-week-old hydrocephalic HTX treated with early shunts were approximately 80% and 40%, respectively. Because of animal death due to shunt malfunction, shunt infection, and subdural hematoma, only 32% of the animals thus treated grew to a sexually mature age and could be subjected to the Y-maze test. The survival rate of 8-week-olds in the delayed shunt group was approximately 20%.

Fig. 1. Ventriculographic follow-up of HTX-rat. Left, non-hydrocephalic HTX-rat with sham-operation (control); Center, hydrocephalic HTX-rat with early shunt insertion at 9 days after birth; Right, hydrocephalic HTX-rat with delayed shunt insertion at 4 weeks after birth

Ventriculography and MRI

The early shunt and delayed shunt groups of hydrocephalic animals displayed distinct differences in the chronological changes in size of their lateral ven­ tricles. In the former group, the size of the lateral ventricle, as assessed by venticulography early during shunt procedure, was abnormally enlarged. However, follow-up ventriculography approximately 4 and 8 weeks after birth revealed that early shunt resulted in normalization not only of enlarged ven­ tricles, but also of cranial vault bulging (Figs. 1 and 2). Normalization of abnormally enlarged ventricles concomitant with increase in cortical mantle thickness following early shunt placement was clearly demonstrated by MRI (Fig. 3). Contrary to these findings, little reduction in the size of abnormally enlarged ventricles was observed in the delayed shunt group (Figs. 1 and 2).

Y-Maze Test: Early Shunt Group

The learning ability of sexually mature HTX whose congenital hydrocephalus had been arrested by an early shunt was assessed by the Y-Maze test and compared with that of sham-operated animals. In both groups, the correct

response rate increased progressively with advancing sessions, as demonstrated by the significant session effect of analysis of variance [F(23,552) = 23.38, P < 0.01 ANOVA]. Although a statistically significant difference of mean correct response could be found in 2 out of 24 sessions, the group effect of analysis of variance [F(1,23) = 1.43 N.S. ANOVA] did not show any statistically significant difference between the two groups. In addition, there was inter­ action between the two factors of group and session [F(23,552) = 1.77, P < 0.05 ANOVA] (Fig. 4a). The response latency time of both groups was pro­ gressively shortened with advancing sessions, as demonstrated by the significant session effect of analysis of variance [F(23,552) = 6.75, P < 0.01 ANOVA].

Comparisons of mean response latency time in each session with Wilcoxon t-test did not demonstrate significant differences between the two groups in all sessions. And then, there was no significant difference between the two groups as demonstrated by group effect of analysis of variance [F(1,23) = 0.06 N.S. ANOVA]. In addition, no interaction between the two factors of group and session was found [F(1,23) = 0.55, N.S. ANOVA] (Fig. 4b).

Fig. 4. Light-darkness discrimination test (Y-maze test): a Comparison of correct response rate between early shunt and sham-operated groups, and b comparison of response latency time between early shunt and sham-operated groups. Vertical bars indicate standard errors of the mean

These findings, obtained from animals in the early shunt group, were found to be entirely different from those of animals in the delayed shunt group, where a significantly lower mean correct response rate and prolonged mean response latency time were observed [F(1,23) = 10.16, P < 0.01, F(1,23) = 10.35, P < 0.01 ANOVA] (Miyazawa and Sato 1991).

Locomotor Activity in Early and Delayed Shunt Groups

When nocturnal locomotor activity of animals in the early shunt group was compared with that in the sham-operated group, there was no remarkable difference in any one hour from 19:00 to 07:00 or in the mean 12-hour cumulative activity counts (Fig. 5). These findings differed from the observation of another study that nocturnal locomotor activity of animals in the delayed shunt group was remarkably greater than that in the sham-operated group (Miyazawa and Sato 1991).

Qualitative and Quantitative Golgi Studies (Takashima et al. 1978)

Histological features such as thickness, length, and degree of bifurcation of the apical and basal dendrites of the pyramidal neurons in layers II and III of the fronto-parietal cortex stained by rapid Golgi technique were found not to be differ among the early and delayed shunt groups and the sham-operated group. However, the spine density of both apical and basal dendrites was remarkably decreased in animals of the delayed shunt group as compared with the density in animals of the respective sham-operated groups (Miyazawa and Sato 1991). On the other hand, the spine density in animals of the early shunt group was not significantly different from that of sham-operated animals (Fig. 6).

Quantitative Histochemical Measurement of SVP-38

Chronological changes in fluorescence intensity in layer I of the parieto­ occipital cortex of hydrocephalic HTX were measured at different times after birth and compared with changes in sham-operated HTX. Fluorescence inten­ sity of the cortex in both hydrocephalic and sham-operated HTX increased linearly up to 3 weeks of age, but after 4 weeks of age, the values for the hydrocephalic rats were significantly decreased in comparison with those of sham-operated animals (Fig. 7). In contrast, the values obtained from early shunt animals were found to be not significantly different from those of sham­ operated animals (Fig. 8).

Immunoblot Analysis of SVP-38

In the cerebrum of 4-week-old non-hydrocephalic HTX, SVP-38 was detected by immunoblot analysis as an intense band with a molecular weight of 38 kilodalton. In contrast, in the cerebrum of 4-week-old hydrocephalic HTX, SVP-38 decayed in parallel with the progression of hydrocephalus. Such decay

Fig. 5. Locomotor activity assessed with Automex II: A Comparison of nocturnal activity between early shunt and sham-operated groups, and B comparison of mean 12- hour cumulative activity between early shunt and sham-operated groups

of SVP-38 was not observed when progression of hydrocephalus had been arrested by early shunt insertion (Fig. 9)

Discussion

In earlier studies, large animals such as cats and rabbits, in which aquired hydrocephalus was induced by such techniques as intracisternal injection of kaolin, were used as hydrocephalic animal models for experimental shunt oper-

Fig. 6. Comparison of the spine density of apical and basal dendrites of pyramidal neurons between early and sham-operated groups in layers II and III of the fronto­ parietal cortex. Horizontal bars indicate standard errors of the mean

Fig. 9. Immunoblot analysis of SVP-38 in the cerebral cortex of 4-week-old HTX-rats. Lanes 1 and 4, non-hydrocephalic HTX-rats with sham-operation; Lanes 2 and 5, hydroceph alic HTX-rats without ventriculoperitoneal shunt; Lanes 3 and 6, hydro­ cephalic HTX-rats with early shunt insertion ation (Granholm L 1966; Hochwald and Epstain 1973; Rubin and Hochwald 1976; Del Bigio and Bruni 1980; Fried et al. 1987). It has, however, become increasingly difficult in recent years to use such large animals in experiments. Such being the case, the use of congenitally hydrocephalic HTX-rats and development of the ventriculoperitoneal shunt technique in small animals appear to be significant in hydrocephalus research. As was reported in the present study, morbidity and mortality of HTX to which the V-P shunt was applied were still very high. Consequently, we felt that techical improvement of the V-P shunt would have to be established before the universal validity of this experimental model could be generally accepted. Nevertheless, various important observations were made in the present investigation.

When chronological changes in size of the lateral ventricles after shunt insertion, assessed by means of ventriculography, were compared in the early and delayed shunt groups, the abnormally enlarged ventricles in the early shunt group were found to be promptly normalized after V-P shunt placement, but this did not occur in the delayed shunt group. That reduction in ventricular size after early shunt placement was accompanied by thickening of the cortical mantle was clearly demonstrated by MRI. These observations may indicate that selection in timing of V-P shunt placement is one crucial factor in regard to prevention of functional and morphological disturbances of the developing brain caused by congenital hydrocephalus in HTX-rats.

The brain weight of a rat reportedly increases rapidly until the 25th day after birth, by which time it may be 85% of the weight of an adult brain. Migration of nerve cells is considered to be a major factor in the brain weight increase during the first 10 days of life. These migrating nerve cells differentiate to produce the cortical cell laminae. During the next 10 days, the axons and dendrites develop and myelin forms. Completion of the neuropil is thought to increase the weight of the cerebral mantle (Haas et al. 1970; Sugita 1971; Wada 1988).

According to quantitative analysis of the morphological development of synapses in the cerebral cortex of rats, the number of synapses increases in linear fashion during the period between 1 and 3 weeks of life, and reachesa plateau thereafter (Aghajanian and Bloom 1967; Armstrong-James and Johnson 1970; Miller et al. 1988; Adams and Jones 1982; Blue and Parnavelas 1983; Markus and Petit 1987). In congenitally hydrocephalic HTX-rats, progressive enlargement of the lateral ventricles, concomitant with thinning of the cortical mantle and periventricular CSF edema takes place in the white matter during a 3-week period after birth (Wada: 1988). Miyazawa et al. (1988), when they investigated cortical pyramidal neurons of such hydrocephalic brains by quali­ tative and quantitative Golgi technique at 2 weeks after birth, found that the neuronal soma remained morphologically intact, although such drastic changes as winding, tortuosity, and varicosity were noted in the apical and basal dendrites, in association with marked reduction in the spine density (Borit and Sidmann 19-72; Marin-Padilla 1972; Purpura et al. 1982; McAllister et al. 1985). Since dendritic spines are now recognized as representing "specific postsynaptic receptive structures" on the dendrites, it was assumed by us that presynaptic structures also may be affected by hydrocephalic pathology. Consequently Miyazawa et al. (unpublished data) demonstrated, by using quantitative histo­ chemical measurement of the synaptic vesicle protein, SVP-38, that in the cerebral cortex of hydrocephalic HTX-rats, a progressive increase in the amount of SVP-38 occurs during a 3-week period after birth, followed by a marked drop in its amount at 4 weeks after birth. The age-related in­ crease of SVP-38 observed in both hydrocephalic and control HTX-rats up to 3 weeks after birth seemed to be reflected by a progressive increase of the synaptic vesicle protein occurring in association with synaptogenesis. In this regard, we felt that the age-related changes of synaptic density, assessed by quantitative electronmicroscopic analysis of synaptogenesis of the rat cerebral cortex, and reported by several investigators such as Aghajanian et a!. (1967), Blue and Parnavelas (1983), and Markus and Petit (1987), lent support to our observation. The sudden decay of SVP-38 4 weeks after birth is considered to be one of the direct effects of the progression of hydrocephalus. This observa­ tion, obtained by quantitative histochemical methods, was also confirmed by immunoblot analysis in the present investigation. Since reduced spine density was noted in the hydrocephalic animals as early as 2 weeks after birth, we speculated that a dissociation of pre- and post-synaptic structures may be present in the disturbance of synaptogenesis.

According to the observations of Jones and Bucknall (1987, 1988), the resting pressure of cerebrospinal fluid in the lateral ventricles of hydrocephalic HTX-rats was not elevated above normal for up to 10 days after birth, but by 21 days the pressure was nearly twice that of normal rats. Miyaoka eta!. (1988) reported that local cerebral glucose utilization (LCGU) of severely affected hydrocephalic HTX-rats was decreased throughout the brain and that applica­ tion of a delayed shunt to such animals at 4 weeks of age could not completely correct the impairment of LCGU. Therefore, the significance of early shunt insertion in these HTX is believed to lie in eradicating the pathological pro­ cesses, which occur secondary to increased intracranial pressure, that affect cerebral blood flow and metabolism in the developing brain. As was demon­ strated in the present investigation, when a ventriculoperitoneal shunt was inserted into animals with progressive hydrocephalus at approximately 4 weeks after birth, not only was there no reduction in size of the abnormally enlarged ventricles, but there was also no increase in cortical mantle thickness. Further­ more, spine density in the cerebral cortex in such animals also was found to be decreased when compared with that in sham-operated animals, and learning disability could not be corrected which differed from results in with sham operated animals. Contrary to these observations, in the present investiga­ tion in HTX-rats we found that insertion of a V-P shunt as early as 7-9 days after birth prevented most of the unfavorable effects inflicted by hydro­ cephalus upon the morphological and biochemical development of the brain. In fact, early shunt placement was found to result in normalization of the abnormally enlarged ventricles, in association with simultaneous thickening of the cerebral cortical mantle and preservation of normal spine density in the dendrites of cortical pyramidal neurons. The learning ability of such animals 24 K. Suda et at.

was not found to be disturbed compared with that of the sham-operated group.

We assume, based on these observations, that early shunt placement may have a beneficial role both in repairing and in preventing the impairment of synaptogenesis which occurs in association with the progression of congenital hydrocephalus. Whether or not there is a close correlation between learning disability and impairment of synaptogenesis, as observed in the experiment with the delayed shunt, is still not clear. However, such a correlation seems to plausible, since the prevention of disturbance of synaptogenesis and lack of learning disability in the early shunt experiment were confirmed in the present investigation.

During human fetal brain development, differentiation of the neocortex extends from the beginning of the 3rd month to the end of the 6th month at gestational age (Tuchmann-Duplessis et al., 1980). Synaptogenesis of the cere­ bral cortex begins from the 7th month of gestation and reaches a maximum at the age of one year (Huttenlocher 1979). Although direct correlation in cerebral development between human and rat is difficult to define, the early and delayed shunts applied to HTX-rats by ourselves may be regarded as a model for treat­ ment of human congenital hydrocephalus in utero and in late infancy. Many investigators have reported that learning disability was found in children whose congenital hydrocephalus was treated during the postnatal period (Milhorat 1972; Dennis et al. 1981; Pretorius et al. 1985). In regard to this, treatment of congenital hydrocephalus in utero, namely, a ventriculoamniotic shunt, was thought to be most promising (Clewell et al. 1982); however the evolution of this new method has been became slowed by the high morbidity and mortality of the fetuses and/or children thus treated (Michejda et al. 1986).

At present, pediatric neurosurgeons generally consider that when congenital hydrocephalus is diagnosed in utero by ultrasound, the affected fetus should be delivered after 33 weeks of gestation and treated by prompt cerebrospinal fluid diversion, when the respiratory distress syndrome can be overcome by medical treatment (Edwards et al. 1986). However, the sequelae of prompt cerebrospinal fluid diversion in premature infants cannot be ignored. Con­ sidering our experimental observations presented here, the development of new treatments for congenital hydrocephalus in utero seems to be one of the tasks required of pediatric neurosurgeons.

References

Adams I, Jones DG (1982) Quantitative ultrastructual changes in rat cortical synapses during early-, mid- and late-adulthood. Brain Res 239: 349-363 Aghajanian GK, Bloom FE (1967) The formation of synaptic junctions in developing rat brain: A quantitative electron microscopic study. Brain Res 6: 716-727 Armstrong-James M, Johnson R (1970) Quantitative studies of postnatal changes in synapses in rat superficial motor cerebral cortex: An electron microscopical study Z Zellforsch 110: 559-568

Blue ME, Parnavelas JG (1983) The formation and maturation of synapses in the visual cortex of the rat. II. Quantitative analysis. J Neurocytol 12: 697-712 Borit A, Sidman RL (1972) New mutant mouse with communicating hydrocephalus and secondary aqueductal stenosis. Acta Neuropathol (Berl) 21: 316-331

Clewell WH, Johnson ML, Meier PR, et a!. (1982) A surgical approach to the treat­ ment of fetal hydrocephalus. N Eng! J Med 306: 1320-1325 Del Bigio MR,. Bruni JE (1980) Changes in periventricular vasculature of rabbit brain following induction of hydrocephalus and after shunting. J Neurosurg 69: 115-120 Dennis M, Fitz CR, Netley CT, Sugar J , Hoffman HJ (1981) The intelligence of hydrocephalic children. Arch Neurol 38: 607-615

Edwards MS (1986) An evaluation of the in utero neurosurgical treatment of ventriculo- megaly. Clin Neurosurg 33: 341-357 ·

Freid A, Schapiro K, Takei F (1987) A laboratory model of shunt-dependent hydro­ cephalus. Development and biochemical characterization. J Neurosurg 66: 734- 740 Granholm L (1966) Induced reversibility of ventricular dilatation on experimental hydrocephalus. Acta Neurol Scand 42: 581-588

Haas RJ, Werner J, Fliedner TM (1970) The cytokinetics of neonatal brain cell development in rats as studied by the complete labeling tritiated thymidine method. J Anat 107: 421-437

Hochwald GM, Epstain F (1973) The relationship of compensated to decompensated hydrocephalus in the cat. J Neurosurg 39: 694-697

Huttenlocher PR (1979) Synaptic density in human frontal cortex. Developmental changes and effects of aging. Brain Res 163: 195-205 Jones HC, Bucknall RM (1987) Changes in cerebrospinal fluid pressure and outflow

from the lateral ventricles during development of congenital hydrocephalus in the HTX Rat. Exp Neurol 98: 573-583 Jones HC, Bucknall RM (1988) Inherited prenatal hydrocephalus in the HTX rat: a morphological study. Neuropathol Appl Neurobiol14: 263-274

Kohn DF, Chinookoswong N, Chou SM (1981) A new model of congenital hydro­ cephalus in the rat. Acta Neuropathol (Berl) 54: 211-218 Kohn DF, Chinookoswong N, Chou SM (1984) Animal models of human disease. Congenital hydrocephalus. Am J Pathol 114: 184-185

Laemmli UK (1970) Cleavage of structual proteins during the assembly of the head of bacteriophage T4. Nature 227: 680-685

Marin-Padilla M (1972) Structual abnormalities of the cerebral cortex in human chromosomal abserrations: a Golgi study. Brain Res 44: 625-629

Markus EJ, Petit TL (1987) Neocortical synaptogenesis, aging, and behavior: lifespan development in the motor-sensory system of the rat. Exp Neurol 96: 262-278

McAllister JP, Maugans TA, Shah MV, eta!. (1985) Neuronal effects of experimentally induced hydrocephalus in newborn rats. J Neurosurg 63: 776-783

Michejda M, Queenan JT, McCullough D (1986) Present status of intrauterine treat­ ment of hydrocephalus and its future. Am J Obstet Gynecol 155: 873-882

Milhorat TH (1972) Hydrocephalus and the cerebrospinal fluid. Williams and Wilkins, Baltimore, pp 177-199

Miller M (1981) Maturation of rat visual cortex. I. A quantitative study of Golgi­ impregnated pyramidal neurons. J Neurocytol 10: 859-878

Millhouse OE (1981) The Golgi methods. In: Heimer L, Robards MJ (eds) Neuro­ anatomical tract-tracing methods. Plenum, New York, pp 311-344

26 K. Suda et a/.

Miyaok M, Ito M, Sato K, Ishii S (1988) Measurement of local cerebral glucose utilization before and after V-P shunt in congenital hydrocephalus in rats. Metab Brain Dis 3: 125-132 Miyazawa T, Sato K (to be published) Learning disability and impairment of synapto­ gene in the HTX-rats with arrested shunt-dependent hydrocephalus. Nerv Syst Child Miyazawa T, Wada M, Sato K (1988) A quantitative Golgi study of cortical pyramidal neurons in congenitally hydrocephalic HTX-rats - Golgi study. Nerv Syst Child 13(4): 263-270

Obata K, Nishiye H, Fujita SC, Shirao T, Uchizono K (1986) Identification of a synaptic vesicle-specific 38000-dalton protein by monoclonal antibodies. Brain Res 375: 37-48

Obata K, Kojima N, Nishiye H, Inoue H, Shirao T, Uchizono K (1987) Four synaptic vesicle-specific proteins: identification by monoclonal antibodies and distribution in the nervous tissue and the adrenal medulla. Brain Res 404: 167-179

Pretorius DH, Davis K, Manco-Johnson ML (1985) Cinical course of fetal hydro­ cephalus: 40 cases. AJNR 144: 827-831

Purpura RC, Bodick N, Suzuki K, Rapin I, Wurzelmann S (1982) Microtubule disarray in cortical dendrites and neurobehavioral failure. I. Golgi and electron microscopic study. Dev Brain Res 5: 287-297

Rubin RC, Hochwald GM (1976) Hydrocephalus III. Reconstitution of the cerebral cortical mantle following ventricular shunting. Surg Neurol5: 179-183

Sholl DA (1953) Dendritic organization in the neurons of the visual and motor cortices of the cat. J Anat 87: 387-406

Sugita N (1972) Comparative studies on the growth of the cerebral cortex. II On the increase in the thickness of the cerebral cortex during the postnatal growth of the brain in the albino rat. J Comp Neurol 28: 511-591

Takashima S, Chan F, Becker LE, Armstrong DL (1978) Morphology of the developing visual cortex of the human infant: A quantitative and qualitative Golgi study J Neuropathol Exp Neurol 39: 487-501

Takiguchi H, lshizuka A, Ikeda Y (1988) The effects of a dibenzoxazepine derivative on learning ability and local cerebral glucose utilization in aged rats. Jpn J Neuro­ psychopharmacol 10: 459-469

Tuchmann-Duplessis H, Auroux M, Haegel P (1980) Illustrated human embryology.

pp 72-74

Wada M (1988) Congenital hydrocephalus in HTX rats: Incidence, pathophysiology and developmental impairment. Neurol Med Chir (Tokyo) 28: 955-964 Zilles K, Wree A (1985) Cortex: A real and laminar structure. In: George P (ed) The rat nervous system, Vol. 1. Academic, pp 375-415