A programmed ependymal denudation precedes congenital hydrocephalus in the hyh mutant mouse.

Hydrocephalic hyh mice are born with moderate hydrocephalus and a normal cerebral aqueduct. At about the fifth postnatal day the aqueduct becomes obliterated and severe hydrocephalus develops. The aim of the present investigation was to investigate the mechanism of this hydrocephalus, probably starting during fetal life when the cerebral aqueduct is still patent. By use of immunocytochemistry and scanning electron microscopy, mutant (n = 54) and normal (n = 61) hyh mouse embryos were studied at various developmental stages to trace the earliest microscopic changes occurring in the brains of embryos becoming hydrocephalic. The primary defect begins at an early developmental stage (E-12) and involves cells lining the brain cavities, which detach following a well-defined temporo-spatial pattern. This ependymal denudation mostly involves the ependyma of the basal plate derivatives. There is a relationship between ependymal denudation and ependymal differentiation evaluated by the expression of vimentin and glial fibrillary acidic protein. The ependymal cells had a normal appearance before and after detachment, suggesting that their separation from the ventricular wall might be due to abnormalities in cell adhesion molecules. The process of detachment of the ventral ependyma, clearly visualized under scanning electron microscope, is almost completed before the onset of hydrocephalus. Furthermore, this ependymal denudation does not lead to aqueductal stenosis during prenatal life. Thus, the rather massive ependymal denudation appears to be the trigger of hydrocephalus in this mutant mouse, raising the question about the mechanism responsible for this hydrocephalus. It seems likely that an uncontrolled bulk flow of brain fluid through the extended areas devoid of ependyma may be responsible for the hydrocephalus developed by the hyh mutant embryos. The defect in these embryos also includes loss of the hindbrain floor plate and a delayed in the expression of Reissner fiber glycoproteins by the subcommissural organ.     

Histological changes in the midbrain around the aqueduct in congenital hydrocephalic rat LEW/Jms

Primary aqueductal stenosis is one of the main causes of congenital hydrocephalus in humans and experimental models. The congenitally hydrocephalic rat strain LEW/Jms is one such model. In this report, we describe further detailed histological features of periaqueductal structure, including the posterior commissure, subcommissural organ (SCO), and ependyma, and discuss the changes in these structures in relation to the cause of hydrocephalus. Coronal sections of the aqueduct in normal rats showed that the usual ependyma was absent in the center of the base facing the dorsal side, which was replaced by tall columnar cells. On the other hand, in hydrocephalic rats the ependyma encircled the aqueductal cavity. In midline sagittal sections, normal and hydrocephalic rats showed the SCO, although the SCO in hydrocephalic rats was shorter than in normal rats. There was also a marked difference between normal and hydrocephalic rats in the dorsoventral dimension of the rostral midbrain. In hydrocephalus, this dimension was large in comparison with normal rats. The superior collicular commissure located caudal to the posterior commissure ran along the ventral side of the midbrain in rats with hydrocephalus, and there was a cell-depleted area just dorsal to the superior collicular commissure. The same findings were observed from the 17th day of gestation until the postnatal period. Although the role of the SCO has been widely discussed from the viewpoint of secretory function, the present study indicated that this organ might be involved in the formation of the shape of the aqueduct.     

Ependymal denudation and alterations of the subventricular zone occur in human fetuses with a moderate communicating hydrocephalus

In mutant rodents, ependymal denudation occurs early in fetal life, preceding the onset of a communicating hydrocephalus, and is a key event in the etiology of this disease. The present investigation was designed to obtain evidence whether or not ependymal denudation occurs in 16- to 40-week-old human fetuses developing a communicating hydrocephalus (n = 8) as compared to fetuses of similar ages with no neuropathologic alterations (n = 15). Sections through the walls of the cerebral aqueduct and lateral ventricles were processed for lectin binding and immunocytochemistry using antibodies against ependyma, astroglia, neuroblasts, and macrophages markers. Anticaveolin was used as a functional marker of the fetal ependyma. The structural and functional molecular markers are differentially expressed throughout the differentiation of the human fetal ependyma. Denudation of the ependyma of the aqueduct and lateral ventricles occurred in all fetuses developing a communicating hydrocephalus, including the youngest ones studied. The denuded surface area increased in parallel with the fetus age. The possibility is advanced that in many or most cases of human fetal hydrocephalus there is a common defect at the ependymal cell lineage leading to ependymal detachment. Evidence was obtained that in hydrocephalic human fetuses a process to repair the denuded areas takes place during the fetal life. In hydrocephalic fetuses, detachment of the ependyma of the lateral ventricles resulted in the (i) loss of the germinal ependymal zone, (ii) disorganization of the subventricular zone and, (iii) abnormal migration of neuroblasts into the ventricular cavity. Thus, detachment of the ependymal layer in hydrocephalic fetuses would not only be associated with the pathogenesis of hydrocephalus but also to abnormal neurogenesis.     

Patterned neuropathologic events occurring in hyh congenital hydrocephalic mutant mice

Hyh mutant mice develop long-lasting hydrocephalus and represent a good model for investigating neuropathologic events associated with hydrocephalus. The study of their brains by use of lectin binding, bromodeoxyuridine labeling, immunochemistry, and scanning electron microscopy revealed that certain events related to hydrocephalus followed a well-defined pattern. A program of neuroepithelium/ependyma denudation was initiated at embryonic day 12 and terminated at the end of the second postnatal week. After the third postnatal week the denuded areas remained permanently devoid of ependyma. In contrast, a selective group of ependymal areas resisted denudation throughout the lifespan. Ependymal denudation triggered neighboring astrocytes to proliferate. These astrocytes expressed particular glial markers and formed a superficial cell layer replacing the lost ependyma. The loss of the neuroepithelium/ependyma layer at specific regions of the ventricular walls and at specific stages of brain development would explain the fact that only certain brain structures had abnormal development. Therefore, commissural axons forming the corpus callosum and the hippocampal commissure displayed abnormalities, whereas those forming the anterior and posterior commissures did not; and the brain cortex was not homogenously affected, with the cingular and frontal cortices being the most altered regions. All of these telencephalic alterations developed at stages when hydrocephalus was not yet patent at the lateral ventricles, indicating that abnormal neural development and hydrocephalus are linked at the etiologic level, rather than the former being a consequence of the latter. All evidence collected on hydrocephalic hyh mutant mice indicates that a primary alteration in the neuroepithelium/ependyma cell lineage triggers both hydrocephalus and abnormalities in telencephalic development.     

New ependymal cells are born postnatally in two discrete regions of the mouse brain and support ventricular enlargement in hydrocephalus

A heterogeneous population of ependymal cells lines the brain ventricles. The evidence about the origin and birth dates of these cell populations is scarce. Furthermore, the possibility that mature ependymal cells are born (ependymogenesis) or self-renewed (ependymal proliferation) postnatally is controversial. The present study was designed to investigate both phenomena in wild-type (wt) and hydrocephalic α-SNAP mutant (hyh) mice at different postnatal stages. In wt mice, proliferating cells in the ventricular zone (VZ) were only found in two distinct regions: the dorsal walls of the third ventricle and Sylvian aqueduct (SA). Most proliferating cells were monociliated and nestin+, likely corresponding to radial glial cells. Postnatal cumulative BrdU-labeling showed that most daughter cells remained in the VZ of both regions and they lost nestin-immunoreactivity. Furthermore, some labeled cells became multiciliated and GLUT-1+, indicating they were ependymal cells born postnatally. Postnatal pulse BrdU-labeling and Ki-67 immunostaining further demonstrated the presence of cycling multiciliated ependymal cells. In hydrocephalic mutants, the dorsal walls of the third ventricle and SA expanded enormously and showed neither ependymal disruption nor ventriculostomies. This phenomenon was sustained by an increased ependymogenesis. Consequently, in addition to the physical and geometrical mechanisms traditionally explaining ventricular enlargement in fetal-onset hydrocephalus, we propose that postnatal ependymogenesis could also play a role. Furthermore, as generation of new ependymal cells during postnatal stages was observed in distinct regions of the ventricular walls, such as the roof of the third ventricle, it may be a key mechanism involved in the development of human type 1 interhemispheric cysts.     

Overexpression of nestin and vimentin in ependymal cells in hydrocephalus

In order to elucidate the immunohistochemical features of hydrocephalic ependyma, immunohistochemical examination was undertaken in 11 normal, post-mortem brains (age range, 11 weeks’ postconception to 6 months after birth) and 12 hydrocephalic brains (three cases each of congenital aqueductal stenosis, Dandy-Walker malformation, Arnold-Chiari type II malformation and posthemorrhagic hydrocephalus) by using antisera to nestin, vimentin and glial fibrillary acidic protein (GFAP). In normal brains, nestin was predominantly expressed in neuroepithelial cells and radial glial fibers during the period of neuronal migration. Vimentin immunoreactivity was principally detected in immature ependymal cells and their basal fibers after the period of neuronal migration, then partly replaced by GFAP reactivity during late gestation. In hydrocephalus, the areas of ependymal disruption were covered with nestin- or vimentin-positive cells. Nestin and vimentin were also expressed in immature ependymal cells or their basal processes in anatomical regions such as the roof or floor plate of the fourth ventricle or the cerebral aqueduct, and the ventral part of the third ventricle. These results suggest that the overexpression of nestin and vimentin in hydrocephalus follows two patterns: a reactive pattern of proliferating immature glial cells associated with ependymal cell loss and an abnormal developmental pattern of immunopositivity associated with anatomical regions in the midline mesencephalon. Received: 27 November 1995 / Accepted: 29 December 1995     

Msx1-deficient mice fail to form prosomere 1 derivatives, subcommissural organ, and posterior commissure and develop hydrocephalus

Msx1 is a regulatory gene involved in epithelio-mesenchymal interactions in limb formation and organogenesis. In the embryonic CNS, the Msx1 gene is expressed along the dorsal midline. Msx1 mutant mice have been obtained by insertion of the nlacZ gene in the Msx1 homeodomain. The most important features of homozygous mutants that we observed were the absence or malformation of the posterior commissure (PC) and of the subcommissural organ (SCO), the collapse of the cerebral aqueduct, and the development of hydrocephalus. Heterozygous mutants developed abnormal PC and reduced SCO, as revealed by specific antibodies against SCO secretory glycoproteins. About one third of the heterozygous mutants also showed hydrocephalus. Other defects displayed by homozygous mutants were ependymal denudation, subventricular cavitations and edema, and underdevelopment of the pineal gland and subfornical organ. Some homozygous mutants developed both SCO and PC, probably as a consequence of genetic redundancy with Msx2. However, these mutants did not show SCO-immunoreactive glycoproteins and displayed obstructive hydrocephalus. This suggests that Msx1 is necessary for the synthesis of SCO glycoproteins, which would then be required for the maintenance of an open aqueduct.     

Immunohistochemical localization of superoxide dismutase in congenital hydrocephalic rat brain

The effects of active oxygen species in the development of congenital hydrocephalus have been investigated. Superoxide dismutase (SOD) is one of the scavengers of active oxygen species and there have been many recent reports on the relationship between neurological disorders by active oxygen species following reperfusion for ischemic brain and SOD. In this study, the localization of Cu-SOD and Zn-SOD in WIC-Hyd congenitally hydrocephalic rat brains was identified by the enzyme unlabeled antibody method. We examined the localization of SOD in the choroid plexus, hippocampus, and ependymal cells of the lateral ventricle and aqueduct of WIC-Hyd rats. SOD was hardly observed in the choroid plexus and faintly localized in the hippocampus and ependymal cells of the congenitally hydrocephalic brain, but was observed equally in the cytoplasm of the choroid plexus, hippocampus, and ependymal cells in control animals. In the hippocampus, less SOD was found in hydrocephalic rats than in controls. The SOD was slightly observed in the CA1 pyramidal cells in hydrocephalic rats. In the lateral ventricle and aqueductal ependyma, less SOD was found in hydrocephalic than in control rats. The amount of Cu, Zn-SOD in the congenitally hydrocephalic rat brain was less than in the control, especially in the choroid plexus. Therefore, we suspect that the production of SOD is congenitally reduced in the congenitally hydrocephalic rat brain, and this may promote the impairment of the function of choroid plexus and cilia due to increased active oxygen species. The reduction of SOD in the choroid plexus, hippocampus and ependymal cells of ventricles or aqueduct may promote the development of hydrocephalus in the congenitally hydrocephalic rat.     

Absence of subcommissural organ in the cerebral aqueduct of congenital hydrocephalus spontaneously occurring in MT/HokIdr mice

Summary The midbrains of pups with congenital hydrocephalus spontaneously occurring in MT/HokIdr mice were histologically examined. The subcommissural organ (SCO) and the posterior commissure were completely absent in the hydrocephalic brain. The cerebral aqueduct in the hydrocephalic brain was never completely stenosed, though it was somewhat narrowed in its middle region as compared with that in the normal brain. a possible interrelationship between an absence of SCO and a cause of congenital hydrocephalus is discussed.     

INHERITED PRENATAL HYDROCEPHALUS IN THE H–Tx RAT: A MORPHOLOGICAL STUDY

The H-Tx rat has inherited hydrocephalus which is present at birth. In order to investigate the onset and early stages of hydrocephalus, the heads of fetuses from 16 to 21 days gestation and at 1 day after birth, were serially sectioned using conventional wax histology. Lateral and third ventricle volumes were measured with a graphics tablet and microcomputer. Hydrocephalus was first detected at 18-20 days gestation by enlarged lateral ventricles and it was sometimes accompanied by a large third ventricle. Most hydrocephalics had a non-patent cerebral aqueduct between the third ventricle and the posterior collicular recess and the remainder (about 25%) had an aqueduct which was patent but with a smaller lumen than in non-hydrocephalic littermates. Some fetuses prior to 18 days gestation with normal lateral ventricles also had non-patent aqueducts. Abnormal aqueducts were lined by ependymal cells which were ventrally displaced by thickening of the overlying midbrain; also the subcommissural organ was foreshortened. Infusion with fluorescent markers confirmed that the flow pathway through the aqueduct was obstructed in many hydrocephalic rats. It is concluded that the hydrocephalus may be the result of abnormal brain development in the midline region of the dorsal mesencephalon, leading to aqueduct closure.     

Central innervation of the rat ependyma and subcommissural organ with special reference to ascending serotoninergic projections from the raphe nuclei

The subcommissural organ (SCO) and the cerebral ependyma receive serotoninergic innervation, but little is known about their origin in the raphe nuclei. Application of the retrograde tracer cholera toxin subunit B (ChB) in the third ventricle resulted in uptake in ependymal axons and backfilling of perikarya in the dorsomedian part of the dorsal raphe nucleus, immediately under the caudal aqueduct. By using dual staining with antisera against serotonin and ChB, a portion of the retrogradely labeled neurons was observed to co-store serotonin. Phaseolus vulgaris–leucoagglutinin (PHA-L) was injected into different raphe nuclei to fill the neurons in the same areas where the retrogradely labeled neurons were found. PHA-L injection in the midline of the dorsal raphe nucleus gave rise to ascending axonal processes in the mesencephalic central gray, from where they entered the periventricular strata and the third ventricular ependyma. In the cerebral ependyma, large numbers of positive fibers were consistently found in the ventral part of the lateral ventricles and in the dorsal part of the third ventricle. A large number of PHA-L-immunoreactive fibers were observed in the hypendymal layer of the lateral part of the SCO. Terminal fibers near the ependymal cells were also observed. In all cases, the PHA-L injections labeled innervating fibers both within the ependyma and in the SCO, whereas injections into the median raphe nucleus or in other raphe nuclei (i.e., the raphe pallidus and the raphe pontis) labeled fibers neither in the SCO nor in the ependyma. This study shows that a specific group of predominantly serotoninergic neurons innervates both the ependyma and the SCO and is probably involved in cerebrospinal fluid regulation. J. Comp. Neurol. 384:556–568, 1997. © 1997 Wiley-Liss, Inc.     

Hydrocephalus with hop gait (hyh): a new mutation on chromosome 7 in the mouse

Hydrocephalus with hop gait (hyh) is a new lethal recessive mouse mutation that arose in the C57BL/10J strain at The Jackson Laboratory. It has been mapped to the proximal end of Chromosome 7 close to the Gpi-1 locus. This homozygous mutant is characterized clinically by a domed head and a hopping gait observable at 2 weeks of age and death between 4 and 10 weeks of age. The affected mice have dilated lateral ventricles and a large third ventricular cyst, patent through narrowed rostral cerebral aqueduct, cystic caudal aqueduct and no communication of the aqueduct with the fourth ventricle. The cerebellum has a mild cortical malformation.     

EXPERIMENTAL OBSTRUCTIVE HYDROCEPHALUS IN THE RAT: A SCANNING ELECTRON MICROSCOPIC STUDY

Hydrocephalus was induced in 12-day old rats by the cisternal infusion of a concentrated kaolin suspension. The animals were killed at day 20 and the ependymal lining of all the ventricles prepared for scanning electron microscopy. The dilation of the ventricles was moderate to gross in all cases. The ependyma of the lateral ventricles was similar in both control and experimental animals. Ependymal damage was present in six out of the twelve hydrocephalic rats. Two had fibres visible on the ependymal surface. Four had tears covered with small round cells, believed to be responsible for the repair of the ependyma. The third ventricle, cerebral aqueduct and fourth ventricle enlarged by incorporating folds of ependyma, present in control animals, into the ventricular walls. The circumventricular organs present in the third and fourth ventricles were not damaged by the dilation of the ventricles, even in severe hydrocephalus.     

Uptake of prolactin from cerebrospinal fluid in rat brain

The fate of prolactin in the cerebrospinal fluid was studied by the use of fluorescein labelled prolactin. The distribution was compared to the immunocytochemical distribution of endogenous prolactin. High intensity fluorescence was found in the ependyma of the area postrema, rostral dorsolateral cerebral aqueduct close to the subcommissural organ, and in some cells of the floor of the cerebral aqueduct. This distribution was not seen when excess unlabelled prolactin was injected. The results suggest prolactin uptake from CSF at specific sites which correspond to sites of localization of immunoreactive prolactin.     

The ependyma in chronic hydrocephalus

H-Tx rats produce congenitally hydrocephalic offspring with varying severity of the condition. We used moderately hydrocephalic rats without evident clinical signs of hydrocephalus and normal controls from the same stock when they were at least 1.5 years old. Macroscopic anatomy was studied by MRI and in fixed brain slices and the ultrastructure of the ependyma, with REM. Apart from markedly stretched areas, where the ependyma was totally destroyed and subependymal structures directly exposed to the CSF, the density of ependymal microvilli and of tufts of cilia was reduced in proportion to the ventricular distension of a given area. A supraependymal “network”– never seen before in acute hydrocephalus – was found, whose purpose is probably to prevent further ventricular enlargement. We conclude that even in arrested hydrocephalus the ependymal sequelae of hydrocephalus are similar to those of the acute stage, illustrating the extremely limited potenctial for recovery, but the organism seems nevertheless to react with an internal stabilization of the ventricular system. Received: 8 January 1998     

Overexpression of nestin and vimentin in the ependyma of spinal cords from hydrocephalic infants

The ependyma of the spinal central canal in cases of hydrocephalus shows abnormalities which vary with the aetiology of ventricular dilatation. To determine whether these ependymal changes are developmental or reactive in nature, immunohistochemical findings were compared between nine normal controls and 12 cases of hydrocephalus (three each of congenital aqueductal stenosis, Dandy‐Walker malformation, Chiari type II malformation, and post‐haemorrhagic hydrocephalus) using antisera to nestin, vimentin and glial fibrillary acidic protein. The main pathological findings were disruption of ependymal layer, apparent pseudostratification of ependyma, expansion, cleft or syrinx formation in relation to the central canal, and ependymal rosette formation. In normal developing fetal spinal cord, nestin and vimentin were expressed mainly in pseudostratified ependymal cells and radial fibres in the median septum. In cases with congenital hydrocephalus (congenital aqueductal stenosis, Dandy‐Walker malformation, and Chiari type II malformation), nestin was overexpressed in immature ependymal cells, and strong vimentin immunoreactivity was detected in the long tract of radial fibres in the median septum. Nestin and vimentin were also expressed in small cells and their fibres which covered areas denuded of ependymal cells in cases of Chiari type II malformation and post‐haemorrhagic hydrocephalus. Two conclusions are suggested by this report. First, the ependyma of the spinal central canal in congenital hydrocephalus shows a delay in maturation of radial glial cells into mature astrocytes and ependymal cells. Second, areas of ependymal denudation may be repaired by the immature glial cells derived from subependymal cells.     

Heterogeneous expression of hydrocephalic phenotype in the hyh mice carrying a point mutation in alpha-SNAP

The hyh mouse carrying a point mutation in the gene encoding for soluble N-ethylmaleimide-sensitive factor (NSF) attachment protein alpha (alpha-SNAP) develops inherited hydrocephalus. The investigation was designed to study: (i) the clinical evolution of hyh mice; (ii) factors other than the alpha-SNAP mutation that may influence the expression of hydrocephalus; (iii) the neuropathological features underlying the different forms of clinical evolution. The study included 3017 mice, 22.4% of which were hydrocephalic. The neuropathological study was performed in 112 mice by use of light and electron microscopy. It was found that maternal- and sex-related factors are involved in the heterogeneous expression of hyh phenotype. The clinical evolution recorded throughout a 4-year period also revealed a heterogeneous expression of the hydrocephalic phenotype. Two subpopulations were distinguished: (i) 70% of mice underwent a rapidly progressive hydrocephalus and died during the first 2 months of life; they presented macrocephaly, extremely large expansion of the ventricles, equilibrium impairment and decreased motor activity. (ii) Mice with slowly progressive hydrocephalus (30%) survived for periods ranging between 2 months and 2 years. They had no or moderate macrocephaly; moderate ventricular dilatation and preserved general motor activity; they all presented spontaneous ventriculostomies communicating the ventricles with the subarachnoid space, indicating that such communications play a key role in the long survival of these mice. The hyh mutant represents an ideal animal model to investigate how do the brain "adapt" to a virtually life-lasting hydrocephalus.     

Topographical distribution of estrogen target cells in the forebrain of platyfish,Xiphophorus maculatus, studied by autoradiography

In the platyfish, Xiphophorus maculatus, after injection of [3H]estradiol-17 beta, nuclear concentration of radioactivity is observed in certain cells of the forebrain. These labeled cells are accumulated in periventricular midline regions that include in the telencephalon the ventral precommissural and dorsal supracommissural areas, and in the diencephalon the preoptic, central hypothalamic and thalamic areas. The specialized ependyma of the subcommissural organ also shows nuclear concentration of radioactivity, while other ependymal cells remain unlabeled. Similar to other vertebrate classes, the accumulations of estrogen target cells exist characteristically in the vicinity of the ventricular system, especially its recesses. This implies close topographical relationships between estrogen target cells and the ventricular recess organs, including the optic recess organ, the infundibular recess organ, the paraventricular organ and the subcommissural organ. Different from other vertebrate classes, no pallial accumulations of estrogen target cells are seen in the teleost.     

Ultrastructure and movement of the ependymal and tracheal cilia in congenitally hydrocephalic WIC-Hyd rats

The aim of the present investigation is to determine whether or not hydrocephalus occurring in hydrocephalic Wistar-Imamichi strain rats (WIC-Hyd) is caused by functional and structural disorders of ependymal cilia. Ultrastructures and movement of cilia in the ependyma of the lateral, III and IV ventricles and aqueduct of Sylvius and in the trachea walls of the animals were examined by means of scanning electron microscopy (SEM), transmission electron microscopy (TEM), and light microscopy using a phase-contrast microscope equipped with a high-speed video recording system. SEM revealed that a marked decrease in the length and number of cilia in the ependymal and tracheal walls occurred in affected male WIC-Hyd. This finding was noted even before the development of ventricular dilatation and was not related to the degree of ventricular enlargement after development of hydrocephalus. A moderate decrease in length and number of cilia was also seen among the normal ciliary tufts in affected female rats which developed a mild degree of hydrocephalus. TEM cilia findings included abnormal axonemal structures such as a lack of dynein arms and displacement of microtubules. The incidence of these ultrastructural abnormalities was found to be greater in affected male rats than in affected female rats. All cilia in affected male rats before and after development of hydrocephalus were immotile. A variety of movement disorders such as immobile, rotatory, and vibratory cilia were observed beside normally beating cilia (motile cilia) in affected female rats which never developed hydrocephalus as severe as that seen in affected male rats. These results seem to indicate that there is a correlation between cilia movement disorder and the degree of ultrastructural abnormalities. Consequently, hydrocephalus developing in affected male and female WIC-Hyd appears to be caused by a motility disorder of ependymal cilia which is part of the primary ciliary dyskinesia (PCD) affecting these animals. The present study appears to indicate that the movement of ependymal cilia may play a role in cerebrospinal fluid circulation, and that dysfunction of ependymal ciliary movement may contribute to development of hydrocephalus in WIC-Hyd rats.