NLRP3 inammasome-mediated cerebrospinal uid hypersecretion in choroid plexus contributes to hydrocephalus after hemorrhage

Background Hydrocephalus is a severe complication of intracerebral hemorrhage with ventricular extension (ICH-IVH). The choroid plexus epithelium plays an important role in cerebrospinal uid (CSF) secretion and constitutes blood-CSF barrier adjusting brain–immune system interface. NLRP3 inammasome is a key component of the innate system which promotes neuroinammation. However, the role of NLRP3 inammasome in the pathogenesis of hydrocephalus after hemorrhage has not been investigated. the mm) was drilled (coordinates:0.2 mm posterior and 2.2 mm lateral to the bregma). A 29-gauge needle was inserted at a rate of 1mm/min at the depth of 5.0 mm from the dura. Using a microinjection pump, 200μl nonheparinized arterial blood was infused into the right caudate nucleus at a rate of 14μl/min through the hole. Normal saline was infused as vehicle groups. NLRP3 (MCE, USA) dissolved in normal saline was administered by intraperitoneal injection 1 hour after ICH-IVH at a dose of 10mg/kg. Bumetanide (MCE, USA) dissolved in normal saline was administered by intraperitoneal injection each day after ICH-IVH at a dose of 10mg/kg.


Abstract Background
Hydrocephalus is a severe complication of intracerebral hemorrhage with ventricular extension (ICH-IVH).
The choroid plexus epithelium plays an important role in cerebrospinal uid (CSF) secretion and constitutes blood-CSF barrier adjusting brain-immune system interface. NLRP3 in ammasome is a key component of the innate system which promotes neuroin ammation. However, the role of NLRP3 in ammasome in the pathogenesis of hydrocephalus after hemorrhage has not been investigated.

Methods
Hydrocephalus after ICH-IVH rat model was accomplished by autologous blood infusion. Then, we investigated the relationship between NLRP3 in ammasome and CSF hypersecretion in choroid plexus.

Results
The NLRP3 in ammasome activated and CSF hypersecretion in choroid plexus epithelium were found after ICH-IVH. NLRP3 inhibition with MCC950 decreased CSF secretion, ventricles dilation and attenuated neurofunction de cits after ICH-IVH. In addition, MCC950 decreased NKCC1 phosphorylation which was the major protein adjusting CSF secretion and improved blood-CSF barrier integrity after ICH-IVH.

Conclusions
This study demonstrates that NLRP3 in ammasome mediated CSF hypersecretion by in uencing NKCC1 phosphorylation in choroid plexus epithelium plays an important role in the pathogenesis of hydrocephalus after hemorrhage and provides a new therapeutic strategy.
Persistent elevations in intracranial pressure can cause acute brainstem herniation and death [5]. Invasive CSF shunting remains the treatment for hydrocephalus after hemorrhage, but which occurred many complications such as shunt obstructions or infections [6,7]. It is necessary to nd targeted pharmacotherapeutic strategies to patients with hydrocephalus [8].
CSF circulation failed is the major mechanism for hydrocephalus after hemorrhage widely accepted. Most of the studies aim to CSF owing out obstacles. After hemorrhage, the ependymal glia and arachnoid granulations dysfunction aggravated hydrocephalus [9,10]. TLR4-dependent in ammatory response contributes to hydrocephalus through CSF hypersecretion [11]. The CSF is predominantly produced by the choroid plexus, an epithelial monolayer which is the main component of the blood-CSF barrier [12,13]. The Na + /K + /2Clcotransporter (NKCC1) expressed in the luminal membrane of choroid plexus contributes approximately half of the CSF production [14]. As one chloride importer antagonist of NKCC1, bumetanide attenuates many neurological and psychiatric disorders. Blood metabolites intracerebroventricular injection caused choroid plexus in ammation and hydrocephalus but the mechanisms remain unclear. Secretory epithelia can respond to proin ammatory stimulation by increasing uid secretion rate.
The NLRP3 in ammasome as a critical component of the innate immune response to tissue injury, which could release cytokines and exacerbated brain edema after ICH. MCC950 is the speci c inhibitor of NLRP3 in ammasome which reduces brain injury after hemorrhage.
However, the pathogenesis of hydrocephalus after hemorrhage still are limited to explain. Therefore, we hypothesized that NLRP3 could aggravate hydrocephalus after ICH-IVH via enhancing CSF secretion in choroid plexus. The present study set out to test the function and molecular mechanism of NLRP3 in ammasome in the pathogenesis of hydrocephalus, then nd a new therapeutic target for hydrocephalus patients.

Animals
Adult male Sprague-Dawley rats (220g-250g;) were purchased from the Army Medical university and kept on a feedback-controlled heating pad to maintain body temperature at 37.0℃. Animal using procedures was in compliance with the guide for the care and use of laboratory animals and approved by the animal care and use committee at the Army Medical university (SCXK-PLA-20120011).

ICH-IVH Model of hydrocephalus and drug treatment
The surgical procedures for rats subjected to hemorrhage were well described in our previous study [9,15]. In brief, animals were anesthetized with pentobarbital (40 mg/kg intraperitoneal). The right femoral artery was catheterized as a source of blood sample. The rats were positioned in a stereotaxic frame, a cranial burr hole (1 mm) was drilled (coordinates:0.2 mm posterior and 2.2 mm lateral to the bregma). A 29gauge needle was inserted at a rate of 1mm/min at the depth of 5.0 mm from the dura. Using a microinjection pump, 200μl nonheparinized arterial blood was infused into the right caudate nucleus at a rate of 14μl/min through the hole. Normal saline was infused as vehicle groups. NLRP3 inhibitor MCC950 (MCE, USA) dissolved in normal saline was administered by intraperitoneal injection 1 hour after ICH-IVH at a dose of 10mg/kg. Bumetanide (MCE, USA) dissolved in normal saline was administered by intraperitoneal injection each day after ICH-IVH at a dose of 10mg/kg.

Brain water content
Three days after blood infusion, rats were euthanized without perfusion. Brains were removed immediately and divided into three parts: left-hemisphere, right-hemisphere and cerebellum. Firstly, the wet weights were measured using fresh tissue. Then, the brain sections were dried at 100℃ for 24 hours before obtaining the dry weights. The percentage brain water content was calculated as follows formula: (wet weight-dry weight)/wet weight100.

Neurobehavioral examination
Neurological dysfunction of rats was evaluated using a modi ed Neurological Severity Score (mNSS) method as described previously [16]. Brie y, the assessment was performed at 3 and 7 days after ICH-IVH. First, the mNSS are a composite test of motor, sensory, and balance functions. Neurological function was graded on a scale of 0-18 (a score of 13-18 indicates severe injury, 7-12 indicates moderate injury, and 1-6 indicates mild injury).

Quali cation of CSF secretion rate
Rates of CSF production were measured using the method as previously described [17]. Brie y, anesthetized rats were mounted on a stereotactic apparatus, and a cranial burr hole (1.2 mm) was drilled over the left lateral ventricle (coordinates, relative to the bregma: x, -0.6cm; y, -1.6cm). Next, the rat's head was rotated on the ear-bars by 90° so that it was oriented nose down, and the suboccipital muscles were dissected to the cisterna magna to expose the atlantooccipital ligament. The ligament was punctured, and a 29-gauge needle was advanced 5 mm through the foramen of Magendie to the fourth ventricle. Sterile molecular-grade oil (100μl; Sigma-Aldrich) was infused through the tube to occlude the aqueduct of sylvius, thereby creating a closed system of CSF circulation in lateral ventricle. With the rat in the same position, a glass capillary tube (OD,1.1mm; ID,1.0mm; length,20cm) was advanced through the burr hole into the lateral ventricle (depth 4.5mm ventral). The volume of CSF that collected at a given time (20 min) was calculated as: V (mm3) =π·r 2 ·d, where r is the radius of the glass capillary tube and d is the distance that CSF traveled within the capillary. The rate of CSF formation (μl/min) could be calculated from the slope of volume-time relationship.

Transmission electron microscope
Electron microscopy was performed as previously described [18]. Rats were anesthetized and subjected to intracardiac perfusion with 4% paraformaldehyde and 2.5% glutaraldehyde in 0.1mol/L Sorensen's buffer (pH 7.4). The choroid plexus were removed from brain and a 1-mm-thick coronal tissue slice was cut with a blade 4 mm overnight at 4°C. Samples were then post-xed with 1.0% OsO4 and dehydrated in graded ethyl alcohol. After completion of dehydration, samples were in ltrated with propylene oxide, embedded in Epon resin, and sectioned. Ultra-thin sections were then stained with uranyl acetate and Reynold's lead citrate. Sections were evaluated using a Philips CM 100 transmission electron microscope (Hillsboro, OR, USA) and digitally acquired using a Hamamatsu (Hamamatsu City, Shizuoka, Japan) ORCA-HR camera.
Blood-CSF Barrier permeability measurement CSF was obtained at 3 days after blood infusion by puncture of the cisterna magna and stored at -80℃ prior to analysis, and dextran-10,000 MW (Invitrogen, USA) was given by vein 2 hours before collecting CSF. CSF was measured absorbance at 540nm using Spectrophotometer.

Cytokines Assay Kits
Rats were anaesthetized with pentobarbital (40 mg/kg intraperitoneal). CSF was collected through the cisterna magna, then rats were decapitated to get the choroid plexus. The CSF was centrifuged at 10,000 rpm to separate the supernatant. The choroid plexus were homogenized in ice-cold PBS with protease inhibitor cocktails (Sigma-Aldrich, USA). Total protein concentration was quanti ed by the BCA protein assay (Boster, China). The CSF and choroid plexus cytokines levels were measured using in ammation cytokines assay kits (Raybiotech, USA) according to the manufacturer's instructions.

Immuno uorescence staining
Immuno uorescence staining of brain tissue was performed on xed frozen sections as previously described. Rats were anesthetized with pentobarbital (40 mg/kg intraperitoneal) and perfused with 4% paraformaldehyde for 4 to 6 hours. The brain was isolated and then dissected the choroid plexus under magni cation using sharp forceps. The choroid plexus were immersed in 30% sucrose for 3 to 4 days at 4℃. The tissues were embedded in an optimal cutting temperature compound (SAKURA, USA), and 12mm-thick slices were cut using a cryostat. The slices were incubated with the following primary antibodies at 4℃ overnight: rabbit antibody to NLRP3(1:200, Abcam, UK), rabbit antibody to p-NKCC1(1:200 dilution, Sigma-Aldrich, USA). Then the slices were probed with appropriate secondary antibodies for 2 h at 37 °C. Finally, the slices were counterstained with DAPI (Boster, China) and examined using a confocal uorescence micro-scope (LSM780, Zeiss).

Cell counts
The cell counts were performed at day 3 after ICH-IVH, respectively. Cell counts analysis was performed as previously described. For quanti cation of the positive cells in the choroid plexus, consecutive slices were made, and two sections per animal (n=6 per group) with 40-μm space in between were used for cell counts. Three high-power images were used for cell counting. NLRP3 and p-NKCC1 positive cells were counted in the choroid plexus. Cell counts were performed by two researchers in a blinded manner. All measurements were repeated three times, and the mean value were used.

MRI and ventricular volume analysis
Rats were anesthetized with 2% iso urane/air mixture throughout MRI examination. The MRI scans were performed in a 7.0-T Varian MR scanner (Bruker, USA) with a T2*gradient-echo sequence and a T2 fasts spin-echo sequence using a view eld of 35mm*35mm and 17 coronal slices (1.0mm thickness). Volumes were calculated as previously described [20]. Bilateral ventricles and hippocampus were outlined, and the areas of all slices and multiplying by the section thickness [21]. All image analyses were performed using Image J (National Institutes of Health, Bethesda, Maryland, USA) by two observers in blinded manner.

Statistical analysis
The values in this study are presented as mean ± SD. SPSS 19.0 software was used to analysis all data. Data were analyzed by Student t test for single comparisons or ANOVA with post hoc Bonferroni-Dunn correction for multiple comparisons. A P value of <0.05 was considered statistically signi cant.

Results
ICH-IVH aroused CSF hypersecretion hydrocephalus and neurocognitive functional de cits 3 days after blood infusion, the frozen sections and T2 MRIs showed ICH-VIH group more severer lateral ventricle dilation than vehicle group (Fig.1A), and using the T2 MRIs images to quantitative measure the lateral ventricle volume got the same conclusion (9.58±1.16 versus 38.17±4.3 mm³; P<0.01; Fig.1C).The brain water content of the ipsilateral hemisphere in the ICH-IVH group was much higher than which in vehicle group(77.97±0.65 versus 78.58±1.32; P<0.01; Fig.1B). The CSF secretion rate of ICH-IVH group was much higher than that in the vehicle group at 3 and 7 days (3 days: 0.93±0.21 versus 2.62±0.53 μl/min; 7 days: 0.42±0.17 versus 0.93±0.14 μl/min; P<0.01; Fig.1D). In addition, the mNSS used to evaluate the neurofunction showed ICH-IVH group had more obvious loss of neurofunction than vehicle group at 3 and 7 days (Fig.1E). We observed hydrocephalus and CSF hypersecretion after ICH-IVH.
NLRP3 in ammasome components were activated in choroid plexus and CSF after ICH-IVH At 3 days after ICH-IVH, we found more NLRP3 positive cells in choroid plexus than vehicle group ( Fig.2A and Fig.2C). In addition, the protein expression of NLRP3, Caspase-1, and IL-1beta of ICH-IVH group were much higher than vehicle group in choroid plexus ( Fig.2B and Fig.2D-2F). Using cytokine assay kits we found that reported NLRP3 related cytokine factors IL-1beta and IL-18 were much higher after ICH-IVH in choroid plexus (Fig.6D and Fig.6E). Compared with vehicle group, IL-1beta and IL-18 also had higher protein expression in CSF (Fig.6A and Fig.6B). According to assay kits results, many in ammation related cytokines increased after ICH-IVH both in choroid plexus and CSF (Fig.S1). These results demonstrated that NLRP3 in ammasome components were activated after ICH-IVH.
MCC950 decreased NKCC1 phosphorylation after ICH-IVH 3 days after ICH-IVH, more p-NKCC1 positive cells were found in ICH-IVH group than vehicle group, and MCC950 treated decreased p-NKCC1 positive cells in choroid plexus ( Fig.4A and Fig.4C). Compared with vehicle group, the protein expression of p-NKCC1 was reduced by MCC950 after ICH-IVH ( Fig.4B and Fig.4D). Based on these, we presume NKCC1 is related to NLRP3 in ammasome.

MCC950 repaired Blood-CSF damages in choroid plexus after ICH-IVH
According to the cytokine assay kits, we found that ICAM-1 higher expressed in ICH-IVH group than vehicle group both in CSF and choroid plexus, and MCC950 treated reduced ICAM-1 protein expression level ( Fig.6C and Fig.6F). Then, the dextran-10,000 MW was used to assess the integrity of blood-CSF barrier. ICH-IVH group had more dextran leakage than vehicle group (Fig.7A), and absorbance at 540 nm also supported this result (Fig.7C). TEM images of choroid plexus showed more tight junctions destroyed in ICH-IVH group than vehicle group (Fig.7B). The protein expression of ZO-1 was much lower after ICH-IVH ( Fig.7D and Fig.7E). After treated with MCC950, the leakage of dextran and CSF absorbance at 540 nm were ameliorated rather than ICH-IVH group (Fig.7A and Fig.7C). Compared with ICH-IVH group, MCC950 treated improved tight junctions in TEM images and the protein expression of ZO-1 (Fig.7B and Fig.7D-E). This result indicated that tight junctions between choroid plexus epithelial cells and blood-CSF barrier were destroyed which may be a source of CSF after ICH-IVH. These damages contributed to aggravate hydrocephalus and MCC950 treated could repair damaged blood-CSF barriers.

Discussion
In the present study, we show that NLRP3 in ammasome was activated which contributed to hydrocephalus by mediating CSF hypersecretion and damaging blood-CSF barrier in choroid plexus after ICH-IVH. What's more, NLRP3 in ammasome in uenced CSF secretion through adjusting NKCC1 phosphorylation in choroid plexus. Together, these results suggest that NLRP3 in ammasome contribute to the pathogenesis of hydrocephalus after ICH-IVH.
The hydrocephalus is a severe complication after ICH especially with ventricles extension [22,23]. Most studies about hydrocephalus after hemorrhage is based on intraventricular hemorrhage (IVH) model no matter in infants or adults which cannot simulate the clinical pathogenesis of hydrocephalus after hemorrhage [24]. As our previous study, hydrocephalus and brain tissue injury are more serious in a rat model of ICH-IVH than IVH, which is more conformer to the clinical [9]. Therefore, we use ICH-IVH model to explore pathogenesis of hydrocephalus in this study.
It is a commonly held view that hydrocephalus after hemorrhage is due to alterations in the CSF drainage pathway, particularly related to the cerebral aqueduct, fourth ventricular outlets, ependymal epithelium damage, arachnoid villi or granulations, and choroid plexus in ammation. There are few animal studies that actually investigate either the cerebral aqueduct or the outlets of the fourth ventricle. The arachnoid granulations are regarded as sites of absorption. Both the presence and morphology of arachnoid granulations is different between humans and animals [25]. However, the arachnoid granulations still are not completely developed in infants hydrocephalus [26,27]. After nerve tissue has lymphatic vessels were widely proved [28][29][30]. It is accepted that a considerable amount of CSF ows into lymphatic vessels and lymphatic out ow may result in hydrocephalus [31]. Beating of the ependymal cilia generates ow of the CSF within the brain cavities and aids in maintaining patency of the ventricular system [32,33]. The clinical and preclinical data suggest that loss of developing and mature ciliate epithelial cells contribute to hydrocephalus after hemorrhage [34][35][36]. All of these studies concentrate on the out ow dysfunction of CSF. The choroid plexus located at the base of each of the four ventricles. As for the produce of CSF in choroid plexus, there are few related studies. Nuclear factor κB (NF-κB) signaling was activated by the CSF barrier cells of the choroid plexus and ependymal lining after IVH [37]. Systemic in ammation stimulates TLRs in the choroid plexus, which may lead to disturbances in choroid plexus barrier function [38]. Recently study proved that TLR4-dependent in ammation leads to CSF hypersecretion plays a role in hydrocephalus after hemorrhage [11]. It has been reported that NLRP3 mediated many kinds of nervous system injuries after ICH, subarachnoid hemorrhage, or traumatic brain injury [16,39], but there are no studies about the NLRP3 in ammasome in hydrocephalus. Based on these, we focus on the NLRP3 in ammasome and choroid plexus, then proved that NLRP3 contributes to hydrocephalus by increasing CSF secretion in choroid plexus. The method in this study to measure CSF secretion is single, and more ways are needed to explore. Furthermore, Blood components are complex, and which components activated NLRP3 in ammasome in choroid plexus is unclear. Many factors, such as iron, hemoglobin, and transforming growth factor-β1, contribute to hydrocephalus after hemorrhage [9,40,41]. Future studies we aim to explore what activates NLRP3 in ammasome after ICH-IVH in choroid plexus.
After we found NLRP3 in ammasome activated after ICH-IVH in choroid plexus, we attempted to nd molecular mechanisms between NLRP3 in ammasome and CSF formation. Approximate 500ml uid is produced by the choroid plexus in the mammalian brain. The CSF production is assumed to take place by transport of osmotically active ions, passive movement of water, and the water channel aquaporin. the Na + /K + /2Clcotransporter 1 (NKCC1) and the K + /Clcotransporters (KCCs) are the main ions transporters adjusting CSF formation, and the NKCC1 plays a decisive role [14,42,43]. The CSF production only declined by a mere 20% in the AQP1 knock-out mice [44], AQP4 also contributes to CSF formation, but its function also limited [45]. The main regulation molecule NKCC1 was selected in this study, and it was brie y proved that NKCC1 is one molecular in NLRP3 mediated CSF hypersecretion. Because of limited evidences, further studies are still needed to clarify NLRP3 mediated CSF hypersecretion after ICH-IVH.
The most widely recognized choroid plexus role is as the site of the blood-CSF barrier, controller of the internal CNS microenvironment [46]. Blood-CSF barrier integrity is impaired in the pathology of many common CNS disorders such as Alzheimer's disease, Parkinson's disease and stroke [47]. Neuroin ammation is a part of the response to CNS injury and disease when blood-CSF barrier dysfunction [48]. We found blood-CSF barrier dysfunction after ICH-IVH, and inhibited NLRP3 in ammasome could improve its integrity. The relationship between blood-CSF barrier integrity and NLRP3 in ammasome is needed to explore.
In this study, we observed that ICH-IVH induced CSF hypersecretion and blood-CSF barrier disruption by activating NLRP3 in ammasome, and the ion cotransporter NKCC1 regulating CSF secretion also participated in this process. In addition, we found lipid droplets formation in the choroid plexus by accident. Lipid droplets (LDs) are the major lipid storage organelles of eukaryotic cells which are central players in anti-infection [49]. Our further studies will focus on what components in blood contribute to NLRP3 in ammasome activation and LDs formation. Although we found a new therapeutic target for hydrocephalus after hemorrhage, but the pathogenesis of hydrocephalus still needed to explore.

Conclusion
Our results showed that NLRP3 activated in choroid plexus contributed to CSF higher-secretion and aggravated hydrocephalus after ICH-IVH, and NLRP3 impacted CSF secretion by regulating NKCC1 phosphorylation level. In addition, blood-CSF barrier permeability damaged by NLRP3 activated in choroid plexus after ICH-IVH, which also contributed to aggravate hydrocephalus. This study provides evidence that inhibiting NLRP3 may be a potential therapeutic for preventing hydrocephalus after ICH-IVH. Figure 1 CSF secretion rate and hydrocephalus evaluation index after ICH-IVH. (A) Representative histological section of brain tissue and T2-weighted images obtained at 3 days after ICH-IVH. (C) Quanti cation volumes of lateral ventricle according to the related T2-weighted images (6 rats/group). (B) Brain water content determination of left, right hemisphere and cerebellum on 3 day after ICH-IVH (6 to 8 rats/group).

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.