Chronic Intraventricular Cannulation for the Study of Glymphatic Transport

Abstract

Glymphatic transport in rodents has primarily been studied using cisterna magna cannulation (CMC), a minimally invasive method for cerebrospinal fluid (CSF) tracers infusion. However, CMC is suboptimal due to the lack of bony structures to stabilize the cannula, leading to potential movement artifacts. Here, we present an alternative approach involving chronic cannulation of the lateral ventricles of mice for CSF tracer delivery. A direct comparison demonstrated that intraventricular cannulation (IVC) reproduces CMC results in vivo, including perivascular labeling of the middle cerebral artery, which was further confirmed by ex vivo analysis. IVC enables tracer infusion in awake mice, facilitating glymphatic transport studies in conjunction with behavioral assessments that were previously unattainable. Additionally, IVC supports repeated infusions in awake animals, offering the potential to reduce the number of experimental animals required. This study establishes IVC as a robust alternative for studying glymphatic transport and associated physiological processes.

Keywords: CSF; clearance; glymphatic flow; glymphatic system; intraventricular cannulation; tracer delivery.

Visual Abstract

Figure 1.

Chronic intraventricular cannula placement. A, 30G needle inserted into PE10 tubing attached to 26G internal cannula. B, The anesthetized mouse is fixed in the stereotaxic frame. The head is leveled using the bite bar holding the nose. C, The dorsal side of the skull is exposed. Bregma and lambda are visible and leveled. D, Starting from Bregma, the coordinates AP −0.6 mm, ML 1.2 mm are located and marked. A hole is drilled in the skull, exposing the brain cortex. E, The cannula is lowered DV −2.0 mm and fixed in place using a mix of glue and dental cement. F, The mouse fully recovers after surgery.

Figure 2.

Intraventricular tracer infusion labels glymphatic flow of cerebrospinal fluid similar to cisterna magna infusion. A, Diagram of the mouse head during in vivo imaging. B, Timeline of the experiment. C, Top, Cisterna magna (CMC) tracer infusion of bovine serum albumin conjugated to Alexa Fluor 647 (BSA-647) and in vivo imaging of dorsal middle cerebral artery (MCA). Bottom, Contralateral in vivo imaging of MCA with an IVC BSA-647 tracer infusion. IVC injected BSA-647 circulates along the brain surface and perivascular spaces of the MCA. D, CMC-infused BSA-647 fluorescence around the MCA increases over time (n = 5). E, IVC CSF tracer fluorescence around the MCA increases over time (n = 4). F, CMC-injected BSA-647 tracer fluorescence rate of change. G, IVC injected BSA-647 tracer fluorescence rate of change. H, Top half, Dorsal (left) and ventral (right) BSA-647 influx in mice with 10 µl IVC infusion after 30 min of circulation time. Bottom half, Dorsal (left) and ventral (right) BSA-647 distribution on mice with 10 µl CMC Infusion after 30 min of circulation time. Scale bar, 2 mm. I, Mean fluorescent intensity (A.U.) of the dorsal (p = 0.0144) and ventral (p = 0.8510) areas of the IVC- and CMC-injected brains (n = 4 IVC, 5 CMC). Groups were compared using unpaired two-tailed t-test. Statistical comparisons are included in the bar plots.

Figure 3.

Adequate surgery recovery time is required for normal glymphatic flow. A, Timelines and representative brain sections for different CSF tracer infusion paradigms. i, Cisterna magna (CMC) with 30 min of circulation time, BSA-647, 10 µl, 2 µl/min. ii, Acute intraventricular (IVC) with 30 min of circulation time, BSA647, 10 µl, 2 µl/min. iii, IVC with 30 min of circulation time, BSA-647, 10 µl, 2 µl/min. iv, IVC with 45 min of circulation time, BSA-647, 10 µl, 2 µl/min. v, IVC with 60 min of circulation time, BSA-647, 10 µl, 2 µl/min. vi, Awake IVC with 60 min of circulation time, BSA-647. B, Mean fluorescent intensity (A.U.) of BSA-647 in the entire brain sections from each injection paradigm. C, Mean fluorescent intensity of cortical BSA-647 from each injection paradigm. CMC, 30 min of circulation time (n = 8); acute IVC, 30 min of circulation time (n = 7); IVC, 30 min of circulation time (n = 9); IVC, 45 min of circulation time (n = 8); IVC 10 µl, 60 min of circulation time (n = 8); awake IVC, 60 min of circulation time (n = 4). Groups were compared using ordinary one-way ANOVA. Statistical comparisons are included in the bar plots.

Figure 4.

Multiple intraventricular infusions can be administered to the same animal. A, Timeline of the experiment. B, Top, Contralateral in vivo imaging at Day 4 of a mouse with an IVC infusion, BSA-TxRed infused and circulated for 60 min. Bottom, Same mouse, contralateral in vivo imaging of BSA-647 infused and circulated for 60 min. C, Mean intensity (A.U.) of BSA-TxRed tracer increases over time, quantified from regions of interest (ROIs) drawn on the middle cerebral artery (MCA; n = 4). D, IVC mean fluorescent intensity (A.U.) of BSA-647 tracer increases over time, quantified from ROIs on the MCA (n = 4). E, Representative brain section of BSA-TxRed tracer distribution at brain extraction. F, Representative slice of BSA-647 tracer distribution at brain extraction.

Figure 5.

Intraventricular cannula implantation may expand ventricles independent of infusion volume and 1 µl infusion volume labels glymphatic influx. A, Representative brain slices +1.20 mm from bregma after being infused with a solution of BSA-647. i, Representative region of interest (ROI) used to measure whole slice mean fluorescence intensity (A.U.). ii, Representative ROI used to measure ventricle size. iii, Representative internal ROI used to exclude ventricular fluorescence. Scale bar, 2 mm. B, Ventricular area quantified from six brain slices of each mouse with different infusion volumes. IVC 10 µl, 2 µl/min, 30 min of circulation time (n = 7). CMC 10 µl, 2 µl/min, 30 min of circulation time (n = 8). IVC 1 µl, 0.2 µl/min, 30 min of circulation time (n = 10). Statistical difference between IVC 10 µl 30 min and CMC 10 µ 30 min. No statistical difference between CMC 10 µl 30 min and IVC 1 µl. C, Mean intensity (A.U.) of whole brain slices from mice infused with 1 µl BSA-647, 30 min of circulation time versus mice infused with 1 µl, BSA-647, 60 min of circulation time. D, Mean intensity (A.U.) of cortical influx brain slices from mice infused with 1 µl BSA-647, 30 min of circulation time versus mice infused with 1 µl, BSA-647, 60 min of circulation time. Groups were compared using ordinary one-way ANOVA. Statistical comparisons are included in the bar plots.

Figure 6.

Iba-1 microglial expression increases in IVC. No difference between microglial Iba-1 expression between surgery site and contralateral side. A, Representative contralateral and ipsilateral brain slices 1.80 mm from bregma from nonsurgical control mice after being labeled for Iba-1 through Immunohistochemistry. B, Representative contralateral and ipsilateral brain slices 1.80 mm from bregma from IVC mice after being labeled for Iba-1 through Immunohistochemistry. C, Ipsilateral (IL) and contralateral (CL) percentage area covered by Iba1 in control (n = 5) versus cannula implanted mice (n = 15). D, Ipsilateral (IL) and contralateral (CL) number of cells labeled for Iba1 between control (n = 5) and cannula implanted mice (n = 15). Number of cells ipsilateral control = 327, contralateral control = 299, ipsilateral implanted 367, contralateral implanted = 371. E, Ipsilateral (IL) and contralateral (CL) average cell size between control (n = 5) and cannula implanted mice (n = 15). F, Ipsilateral (IL) and contralateral (CL) area covered (in pixels) by Iba1 in control (n = 5) versus cannula implanted mice (n = 15). Groups were compared using mixed effects analysis. Statistical comparisons are included in the bar plots. Scale bar, 500 µm.

Figure 7.

Common surgical mistakes prevent normal tracer influx. A, Representative −1.2 mm from bregma brain slice, dorsal and ventral view of a brain with IVC that did not receive a full infusion of BSA-647. B, Representative −1.2 mm from bregma brain slice, dorsal and ventral view of a brain with IVC in which the cannula did not reach the ventricle. The BSA-647 tracer was instead infused into the brain cortex. C, Representative −1.2 mm from bregma brain slice, dorsal and ventral view of a brain with IVC in which the cannula was inserted in the hippocampus instead of the ventricle. Scale bar, 2 mm.