The experimental findings using the kaolin-induced hydrocephalus rat model demonstrate how chronic hydrocephalus can disrupt the dynamics of CSF circulation associated with the glymphatic system and cerebral lymphatic drainage. It is hypothesized that the injection of kaolin into the rat cisterna magna, which limits CSF absorption through the subarachnoid space, could serve as a reliable model for studying clinical conditions such as chronic hydrocephalus or NPH, a communicating type of hydrocephalus. Although we acknowledge that the kaolin-induced hydrocephalus model, in which the CSF absorption pathway is artificially obstructed by injected kaolin, may not perfectly mimic the pathophysioloy and clinical features of chronic hydrocephalus or NPH, we have modified our experimental settings to resemble chronic hydrocephalus or NPH observed in humans. The aim was to create a chronic and mild condition similar to NPH without directly obstructing the intraventricular circulation of CSF. To achieve this, we used a minimal amount of kaolin and prolonged the duration of the hydrocephalus.
Injection of kaolin into the cisterna magna is known to induce inflammation, which leads to meningeal fibrosis, specifically chemical arachnoiditis [35–38]. The majority of the kaolin deposits were distributed throughout the basal subarachnoid space, evenly on both sides of the ventral brainstem, including the interpeduncular fossa to the medulla and pons. No significant presence of kaolin was observed within the intraventricular system, particularly blocking the outlet of the fourth ventricle. The icv-injected EB diffused throughout the entire subarachnoid space through the outlet of the fourth ventricle. This diffusion pattern of EB indicates that CSF flow through the intraventricular and subarachnoid spaces was not mechanically obstructed in the kaolin-induced chronic hydrocephalus rat model.
When discussing hydrocephalus, it is important to differentiate between acute and chronic conditions based on duration. The kaolin-induced hydrocephalus model is a well-known animal model for chronic hydrocephalus as it develops and progresses hydrocephalus over several weeks to months [21, 35, 39–41]. Injected kaolin in the basal subarachnoid space can cause acute hydrocephalic symptoms within two weeks, leading to an increase in resistance to CSF outflow and subsequent elevation of intracranial pressure [35, 39]. As hydrocephalus progresses from the acute phase to the chronic phase over a longer period of four to six weeks or more, the resistance to CSF outflow and intracranial pressure undergo changes [35, 39]. Animal studies using MRI have shown ventricular enlargement, with a peak at 6 weeks, followed by a subsequent decline at 10 weeks [21]. The progressive increase in ventricular size up to 8 weeks reaches a new steady state characterized by the restoration of normal intracranial pressure and an increase in resistance to CSF outflow [42]. Moreover, the normalization of cerebral blood flow after 8 weeks indicates that the post-operative 8-week time point we chose for evaluation is optimal for investigating chronic hydrocephalus, such as NPH [40]. Our model also identified gait impairments, which are clinical manifestations observed in human NPH [6]. Although the kaolin-induced hydrocephalus rat model has inherent limitations, it has potential for investigating the impact of chronic hydrocephalus on the glymphatic system and cerebral lymphatic drainage in neurodegenerative diseases, particularly in relation to the accumulation of metabolic waste.
The kaolin-induced hydrocephalus rat model’s pathophysiology is mainly focused on neuroinflammatory changes and alterations of AQP4, the brain’s most abundant water channel that regulates water homeostasis [43]. Neuroinflammation with gliosis in the basal forebrain and corpus callosum may be indirectly triggered by increased pressure in the subarachnoid or intraventricular space, in addition to the reactive gliosis due to direct contact with kaolin. The degree of gliosis is strongly correlated with enlarged ventricular volume, which may indirectly support the possibility of pressure-related gliosis. A significant depolarization of AQP4 was observed, resulting in the translocation of the normal perivascular pattern of AQP4 to the parenchymal pattern with an increased total AQP4 signal in the CH group. The evidence indicates that neuroinflammation and depolarization of AQP4 play a significant role in the pathophysiology of impaired CSF circulation in chronic hydrocephalus.
Previous studies have reported impaired glymphatic system associated with the depolarization of AQP4 [25, 26]. In this study, we investigated the glymphatic efflux of intraparenchymally injected tracers with two different molecular weights. As anticipated, the molecular weight-dependent intraparenchymal dispersion of the tracers at 6 hours post-injection was hindered in the CH group, indicating a compromised glymphatic system in the chronic hydrocephalus.
Previous studies have evaluated the classical CSF circulation pathway, which involves production in the choroid plexus, intraventricular circulation, and absorption into the dural venous sinuses through the arachnoid villi [7, 44]. To assess this pathway, we measured the subarachnoid dispersion of icv-injected EB. Analysis of the subarachnoid brain surface area covered by EB in the sham group revealed rapid subarachnoid dispersion at 3 hours post-injection, which was subsequently eliminated by 24 hours post-injection. In contrast, the CH group exhibited delayed and stagnated dispersion into the subarachnoid space, indicating a disturbance of the classical CSF circulation pathway in the chronic hydrocephalus.
Furthermore, we investigated an alternative transependymal pathway for CSF circulation. This pathway was discovered in a study that utilized precise MRI mapping of CSF flow with a tracer injected into the ventricle of healthy rats [45]. The investigation identified previously unrecognized parenchymal perivascular space connections that spread across various brain regions, facilitating the direct transport of CSF from the ventricles to the subarachnoid space [45]. In studies of hydrocephalus in humans, the absorption of CSF by periventricular tissues serves as a compensatory mechanism for increased intracranial pressure and sheds light on an alternative pathway for CSF circulation [46–48]. Consistent with these studies, our analysis of coronal brain slices stained by EB revealed rapid periventricular diffusion of icv-injected EB at 3 hours post-injection and subsequent elimination at 24 hours post-injection in the sham group, indicating active transependymal CSF flow in the normal condition. Conversely, minimal periventricular diffusion of EB was observed in the CH group. Based on the confocal microscopic image of the corpus callosum in the sham group, the capillaries were completely filled with EB three hours after injection and showed clear drainage within 24 hours after injection. The microvascular density was found to be deceased in the CH group compared to the sham group, which is consistent with previous studies reporting a decrease in microvascular density in chronic hydrocephalus [49–51]. In the CH group, there was a decrease in capillary density and no visible diffusion of EB into the capillaries even after 24 hours. These findings suggest a disturbance of the alternative transependymal CSF circulation pathway in chronic hydrocephalus, as evidenced by compromised periventricular diffusion and capillary drainage of EB.
In the animal study, the final step of CSF efflux into the venous sinuses through the arachnoid villi was demonstrated using serial high resolution MRI tracking of injected tracer into the lateral ventricles [45]. Time-dependent EB drainage through the venous sinus was observed in the sham group, supporting the classical CSF circulation pathway via venous sinus drainage. Analysis of EB staining on the meninges showed strong staining around the venous sinuses and the middle meningeal artery.
Recent research has also revealed the presence of lymphatic vessels in the meninges responsible for draining CSF, interstitial fluid, macromolecules, and immune cells to the cervical lymph nodes [52, 53]. In a previous study using novel lymphatic reporter rats, it was demonstrated that meningeal lymphatic vessels were located alongside the middle meningeal artery, superior sagittal sinus, and transverse sinuses [54]. In our study, we used a whole-mount dissection of the dura mater and found that LYVE1-positive meningeal lymphatic vessels appeared as discontinuous dotted lines scattered along the COLIV-positive meningeal arteries. Cross-sectional analysis of multi-immunofluorescence signals revealed intraluminal filling and subsequent drainage of EB through the meningeal lymphatic vessels as a CSF efflux pathway. In the CH group compared to the sham group, the EB signal did not appear within the meningeal lymphatic vessels, although the structure of the meningeal lymphatic vessels seemed intact. This suggests a diminished functional CSF efflux through the meningeal lymphatic vessels in chronic hydrocephalus. As expected, peripheral lymphatic drainage to the deep cervical lymph nodes via the meningeal lymphatic vessels or other possible routes was also delayed in the CH group.