Animals

Male and female SD rats aged 3, 8, and 16 months old (MO) (equivalent to ~ 20 (young adult), 45 (midlife), and 60 (aged) human years of age, respectively) were acquired from Envigo (Indianapolis, IN, USA) as previously published by our laboratory [5]. Rats were pair housed until surgical intervention or body weight requirements were exceeded in a 12:12-h light–dark cycle, temperature (20–26 °C) and humidity-controlled (30–70%) environment. Rats were allotted tap water and standard irradiated rodent diet (Envigo Teklab, WI, Teklab Global Diet #2918, 18% protein, 5% crude fat, 5% fiber, total NaCl content 0.6% [174 mEq Na + /kg]) ad libitum. All animal protocols used were approved by the Institutional Animal Care and Use Committee at Boston University Chobanian and Avedisian School of Medicine in agreement with guidelines set by the university and the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Every possible step was taken to minimize pain and suffering. Euthanasia was performed as per the above guidelines, in accordance with the approved IACUC protocol number PROTO201800201 at Boston University.

Surgical proceduresAcute femoral artery and femoral vein cannulation

Male and female SD rats (N = 5–6/group) were anesthetized using sodium methohexital (20 mg/kg intraperitoneally (i.p.), with 10 mg/kg administered intravenously (i.v.) as needed). Cannulation of the femoral artery and vein was performed as previously described [5, 19]. An incision was made in the left femoral triangle and the femoral artery and vein were dissected from the adjacent tissue. Using a cannula made from PE-50 tubing, the cannula was inserted into the femoral vein to allow for administration of i.v. anesthesia, isotonic saline during recovery periods, as well as hexamethonium administration. A PE-50 cannula was placed in the left femoral artery to record heart rate and mean arterial pressure (MAP). Cannulas were tied in place using sutures and the incision was closed. Rats were positioned in a Plexiglas rat holder and the cannula for the femoral artery was attached to an external pressure transducer, while the cannula for the femoral vein was attached to an infusion pump. A two-hour recovery period was allowed, during which an i.v. infusion of isotonic saline (20 μL/min) was performed and rats returned to full consciousness, with stable cardiovascular and renal function. After the two-hour recovery period, baseline MAP was continuously recorded over a one-hour period in conscious rats through the femoral artery cannula using the computer-driven data acquisition software, MP150 and AcqKnowledge 3.8.2 (BIOPAC, CA).

Intravascular infusion of dextrans

Male and female SD rats (N = 6/group) were anesthetized (30 mg/kg ketamine i.p., 3 mg/kg xylazine i.p.) [20] and an incision was made into the left carotid triangle. Muscles were mobilized to expose the carotid sheath. The carotid artery was dissected from surrounding tissues and isolated. A PE-50 cannula was placed into the exposed left common carotid artery and secured [8]. Rats were infused with fluorescein isothiocyanate-Dextran (FITC) at 10 kDa (10 mg/mL, FD10S, Sigma-Aldrich) and rhodamine B isothiocyanate-Dextran at 70 kDa (10 mg/mL, R9379, Sigma-Aldrich) over 60 s (~ 10 μL/1 s). Fluorescent dextrans were allowed to circulate for 20 min before animals were sacrificed via decapitation. An additional group of 3MO male rats (N = 6) were infused with hypertonic mannitol (1.4 mol/L, 2 mL per 200–250 g) 5 min before the dextran infusion as a positive control (Supplemental Fig. 1) [8]. After sacrifice, brain tissue was extracted and post-fixed in 4% paraformaldehyde (PFA) for 48–72 h, then immersed in 30% sucrose solution until sectioning.

Osmotic mini pump insertion

Male rats aged 16MO (N = 5–6/group) were treated with either losartan potassium (LOS; 3 mg/kg/day in sterile saline for 21 days; L0232, TCI) [19] or hydrochlorothiazide (HCTZ; 4 mg/kg/day in 50:50 DMSO:Saline for 14 days; H4759, Sigma) [5] using 2ML4 pumps (Alzet). In brief, rats were anesthetized using a ketamine/xylazine cocktail (30 mg/kg i.p. ketamine; 3 mg/kg i.p. xylazine), an incision was made in the subscapular region and a subcutaneous pocket was made. An osmotic mini pump was implanted, and the incision was closed with suture.

Transcardiac perfusion

Male and female SD rats were anesthetized with a ketamine/xylazine cocktail (30 mg/kg ketamine i.p., 3 mg/kg xylazine i.p.). Once anesthetic depth was reached and there was an absence of toe pinch response, the thoracic cavity was opened and a needle was placed in the heart attached to a perfusion pump. Rats were perfused with ice-cold 1X phosphate-buffered saline before perfusion with 4% PFA. Once fixation was achieved, perfusion was stopped and brains was harvested. Brains were post-fixed in PFA for 48–72 h before being moved to 30% sucrose for sectioning.

In vivo studiesAssessment of vascular sympathetic tone

In animals that underwent acute femoral artery and vein cannulation, peak depressor response to ganglionic blockade (hexamethonium, 30 mg/kg, i.v., H2138, Sigma-Aldrich) was used to estimate sympathetic tone to the vasculature as previously described [5, 19]. Baseline blood pressure was defined as the average blood pressure that occurred during the ten minutes before the bolus of hexamethonium was administered. The lowest blood pressure measurement that occurred within five minutes of the administration of hexamethonium was used to calculate the difference compared with baseline blood pressure.

Novel object recognition task

For novel object recognition (NOR) testing [21], rats were placed in a 60 by 60 cm open topped behavioral testing arena with a video camera fixed above to record testing for manual analysis. Two different objects were placed in separate quadrants of the arena placed 6 inches from adjacent corners. For familiarization, rats were placed in the center of the arena and allowed to explore the environment and objects for 10 min. Following familiarization, rats were returned to their home cages for 2 h. Following the 2-h latency, rats were returned to the testing arena, this time with a novel object replacing one of the familiar objects. Rats were allowed to explore the environment and objects for 10 min to complete the test portion of the assessment. After testing rats were returned to their home cages. The testing arena and objects were thoroughly cleaned between each test with 10% ethanol.

Object location task

Object location (OLT) testing [22] was completed in the same open topped behavioral testing arena as described in the Novel Object Recognition Task. Familiarization was carried out in the same manner. Following the 2-h latency, rats were returned to the testing arena, this time with one of the familiar objects placed in the opposite quadrant from its original location. Following 10 min of exploration, testing was complete and rats were returned to their home cages and the arena was cleaned with 10% ethanol.

In 16MO male rats treated with losartan or hydrochlorothiazide, novel object recognition task and object location task were completed at baseline prior to the initiation of treatment and at the end of the treatment period in order to assess the impact of treatment on cognitive performance in a within subject design.

The time a rat spent investigating each object separately was measured in order to calculate time investigating the novel object or location and the total time spent investigating both objects. These measures were used to calculate the familiarization index, discrimination index, and the percent of time exploring the novel object/location.

Familiarization score was calculated as the absolute value of the time spent at one object subtracted by the time spent at the other object divided by the total time spent investigating the objects. This score was used to assess any preference a rat may have for one object over the other.

$$Familiarization=\frac$$

Discrimination index is a measure of sensitivity and is assessed as the time spent investigating the novel object/location minus the time spent investigating the familiar object/location divided by the total time spent investigating both objects/locations. Percent of time exploring the novel object/location was assessed time spent investigating the novel object/location divided by the total amount of time spent investigating.

$$Discrimination\, index=\frac$$

Molecular techniquesAssessment of plasma norepinephrine content and plasma estradiol content

Plasma taken from male and female SD rats following conscious decapitation was used to measure plasma norepinephrine (NE) via ELISA (IB89552, IBL America) according to the manufacturer’s directions as previously performed by our laboratory [5, 11]. To measure plasma estradiol, plasma taken from female rats following conscious decapitation was used to measure 17β estradiol via ELISA (IB79329, IBL America) according to the manufacturer’s directions as previously performed by our laboratory [5].

Assessment of plasma progesterone content

Plasma progesterone was measured at the University of Mississippi Medical Center Mass Spectrometry Core using LC–MS/MS analysis. All reagents used were high throughput liquid chromatography grade and purchased from Millipore Sigma including water, methanol, acetonitrile, progesterone materials, and isotope-labeled internal standard for progesterone (Progesterone-D9). Cold-induced phase separation was used to extract progesterone from plasma. Briefly, 100 µL of plasma and 10 µL of Internal standard solution (100 ng/mL isotype-labeled internal standards) were added to 200 µL cold acetonitrile and vortexed before an additional 100 µL HPLC water was added. The samples were centrifuged (10,000 g, 2 min, 4℃) and stored at − 20℃ for 30 min to induce phase separation. After the first incubation, the upper layer (100 µL) was transferred to a separate tube containing 100 µL NaHCO3 and then incubated 10 min at 65℃. The samples were then placed back into − 20℃ for 45 min for phase separation. After 45 min, the upper phase (50 µL) was added to 50 µl HPLC water and used for LC–MS/MS analysis. Equipment for detection consisted of an Eksigent M5 microLC with autosampler coupled to a Sciex Qtrap 7500 mass spectrometer. For LC separation, Formic acid in water (0.1% v/v) and formic acid in 75% acetonitrile (0.1% v/v) were used as mobile phases A and B, respectively. The gradient was set as 0–1 min 20–40% B, 1–4.5 min 40–95% B, 4.5–5 min 95% B, 5–5.5 min 95–2%B, and 5.5–8 min 2% B. A Phenomenex Luna 5 µm C18 100 Å (150 × 0.3 mm) LC column was used. The flow rate and column temperature are set at 10 µL/min and 30℃ with an injection volume of 5 µL. The electrospray ionization parameters are as follows: curtain gas (40.0 psi), CAD gas (medium), ionspray voltage (4000 V), source temperature (350 °C), ion source gas 1 (30.0 psi), ion source gas 2 (70.0 psi), and Q0 dissociation (40 V). The multiple reaction monitoring transitions used for quantification were as follows: progesterone, 315.2 → 97.0, and progesterone-D9, 324.2 → 99.9. All peaks were automatically integrated via SCIEX OS AutoPeak integration algorithm and analyzed in Sciex OS v.3.3.

Vascular integrity of the BBB

Following intravascular infusions of dextrans, post-fixed tissue was sectioned at 40 µm and cryoprotected. PVN tissue was selected between bregma − 1.6 and − 2.16 mm, washed in phosphate-buffered saline (PBS) and mounted on gelatin-subbed slides. Coverslips were secured using Prolong Diamond Antifade Mountant with DAPI (P36971, Invitrogen). Analysis is described below.

Immunohistochemistry

Immunohistochemistry was performed as previously described by our laboratory [5, 11]. Brain tissue from perfused rats was sectioned at 40 μm on a cryostat and cryoprotected. Multiple sections sampling all levels of the PVN were selected using Paxinos & Watson’s The Rat Brain atlas, between bregma − 1.6 mm and bregma − 2.16. For immunohistochemistry, sections were washed in 0.1 M PBS 3 times for 10 min each and then incubated in 0.3% hydrogen peroxide for 30 min. Sections were then blocked and permeabilized in PBS-diluent (0.01 M PBS with 3% normal horse serum and 0.25% Triton X-100) for 2 h before incubation in primary antibodies mouse OX-42/CD11b/c for microglia [11] or mouse glial fibrillary acidic protein (GFAP) for astrocytes [23] for 2 h at room temperature and then incubated overnight at 4℃. All antibody information can be found in Supplemental Table 1. Following incubation in primary antibodies, sections were washed in PBS then incubated in biotinylated goat anti-mouse IgG (1:100, BA-9200, Vector Laboratories). Sections were washed again in PBS and incubated in avidin and biotin. Sections were developed in 0.04% 3,3′-diaminobenzidene and 0.04% nickel ammonium sulfate in 0.1 M PBS, mounted on gelatin-subbed slides and dehydrated overnight. Coverslips were secured using Permount mounting medium. Negative controls were performed. One control was performed with the omission of the primary antibody and the inclusion of the secondary antibody, while another control was performed with the inclusion of the primary antibody and the omission of the secondary antibody (Supplemental Figs. 2–4).

Immunofluorescence

For immunofluorescence, PVN sections were selected as described above. Tissue was washed in PBS 3 times for 10 min each and blocked in PBS-diluent for 2 h (0.01 M PBS with 3% normal goat serum and 0.25% Triton X-100). Sections were then incubated in primary antibodies for mouse interleukin 6 (IL-6) [24] or mouse tumor necrosis factor-⍺ (TNF-⍺) [25] at room temperature for 2 h and then incubated overnight at 4℃. All antibody information can be found in Supplemental Table 1. Tissue was brought to room temperature and washed 3 times for 10 min each and incubated in appropriate secondary antibodies for goat anti-mouse AlexaFluor 594 (1:500, A-11005, Invitrogen) for 2 h. At the end of the incubation, sections were washed 3 times for 10 min each and mounted on gelatin-subbed slides and allowed to dry. Coverslips were secured using Prolong Diamond Antifade Mountant with DAPI counterstain. Negative controls were performed. One control was performed with the omission of the primary antibody and the inclusion of the secondary antibody, while another control was performed with the inclusion of the primary antibody and the omission of the secondary antibody (Supplemental Fig. 2–4).

Microscopy and image analysis

Sections were imaged on a Keyence BZ-9000 Fluorescence Microscope with brightfield capabilities. Images were captured at 10X, 20X, and 40X magnifications. Dichroic filters for Texas Red at 585 nm, GFP at 495 nm, and DAPI at 400 nm were used. A representative schematic of PVN neuroanatomy at 10X is included in Supplemental Fig. 5.

For assessment of BBB integrity, 20X images of FITC and Rhodamine were obtained and opened in FIJI/ImageJ. The images were converted to 16-bit, and the colocalization threshold tool was used to isolate matching signal in both images. Images were converted to binary, and the FITC image was subtracted from the colocalization image in order to expose FITC only signal. The percent area of FITC extravasation was calculated by using the threshold tool to select all areas positive for signal within the PVN.

For microglia analysis, images of the left and right PVN were taken at 40X magnification in order to assess morphology and perform Sholl analysis as previously described by our laboratory [11]. A 200 µm by 200 µm box was randomly selected and used for analysis. Microglia were deemed active if a majority of the process lengths were less than that of the diameter of the soma. The number of active microglia and the total number of active microglia were measured. For Sholl analysis, 8 individual microglia from each 40X image were selected at random and cropped for isolation using FIJI/ImageJ. The area of the individual microglia was selected using the threshold tool and the center of the specific randomly selected microglia was identified. Concentric rings 1 µm apart were placed on the image and the number of branching intersections for each randomly selected microglia were calculated (Supplemental Fig. 6).

For astrocytes (GFAP), IL-6, and TNF-⍺ analysis, 20X images of the left and right PVN were used to assess protein expression as total percent area. Images were converted to 8-bit, and positive areas were selected using the threshold tool, and particle analysis was performed.

Statistical analysis

All data are presented as mean ± standard deviation with p < 0.05 as significance. Pearson’s r correlation was performed to assess correlation with age and blood pressure as variables. A one-way ANOVA was used to assess group differences for all analyses with age as a set variable. Post-hoc analysis was performed using a Tukey test. For behavioral comparison between 16 MO male rats and LOS or HCTZ treatment, a repeated measures t-test was used. Prism 9 (GraphPad Software, CA) was used to carry out statistical analysis.