Deconvolved Widefield Fluorescence Microscopy Image Dataset of Carotid Body Chemoreceptors in Wistar and Spontaneously Hypertensive Rats. Part B

PART A of the dataset is available via doi: 10.5281/zenodo.18489925

Image dataset of the neurovascular interface of the carotid body (CB) arterial chemoreceptors. CB from normotensive (Wistar) and Spontaneously Hypertensive rats (SHR) of multiple age groups were serially sectioned, capturing the entire organ, and labelled using markers of chemosensory glomus cells (TH), vasculature (TL/LEL) and innervation (160 kDa Nf-M). Generated multichannel tile images (N=1655) were deconvolded using the Richardson-Lucy algorithm. Dataset generated is meant to support modelling of arterial chemosensory function.

Animals

A total of 48 aged-matched, male Wistar (n=22) and Spontaneously Hypertensive Rats (SHR; n=26) were used in the study:

• 4-7 weeks old (young) Wistar: n=8
• 4-7 weeks old (young) SHR: n=8
• 10-12 weeks old (aged) Wistar: n=8
• 10-12 weeks old (aged) SHR: n=10
• 32-47 weeks ol (old) Wistar: n=6
• 32-47 weeks ol (old) SHR: n=8

Sample collection

Animals were deeply anaesthetized with isoflurane (5% in O₂, 1 L min,−1 via inhalation) until loss of paw withdrawal reflex, then given heparin intraperitoneally (350 UI, Pfizer, Australia). Maintaining isoflurane anaesthesia animals were injected with a lethal dose of Pentobarbital Sodium (300 mg/kg; Euthatal) intraperitoneally and the onset of death was confirmed by absence of withdrawal reflex to a hind paw noxious pinch and respiratory cessation. Animals were intracardially perfused with 300mL of ice-cold phosphate buffered saline (PBS; pH 7.4) and then 300mL of ice-cold 4% (w/v) paraformaldehyde solution (PFA; Sigma-Aldrich) in PBS at a constant rate (~30mL/min) using a peristaltic pump.

Animals were then transferred to a surgical area and common carotid artery bifurcations were accessed via the anterior midline neck incision. The common carotid artery bifurcation was gently separated from the surrounding tissues and cut above the hypoglossal nerve transversing the bifurcation to ascertain intact CB collection. Dissected bifurcations were quickly rinsed in ice-cold PBS to remove any remaining blood and transferred to chilled 4% (w/v) PFA solution. Samples were fixed overnight at 4^oC before transferred to 30% sucrose (S9378; Sigma-Aldrich) solution in PBS and kept for 24 hours at 4°C. Following tissue fixation and cryoprotection, tissue samples were embedded in OCT compound (Shandon™ Cryomatrix™, Thermo Fisher Scientific), frozen on dry-ice, and stored at -80^oC until processed.

Common carotid artery bifurcations were cryosectioned to capture the entire carotid body (CB) in transverse orientation in 20 μm-thick serial sections from rostral to caudal direction. This was achieved by inspecting for the presence of the CB tissue under a benchtop brightfield microscope where a ‘pilot’ section was examined every 100 μm after staining with 0.1% (w/v) toluidine blue (89640; Sigma-Aldrich). Once the first glomus cell clusters were identified all subsequent tissue sections were mounted on SuperFrost Plus™ (10149870; Thermo Scientific) microscopy slides. The collection was stopped once the absence of the CB was confirmed by contrast illumination under the benchtop brightfield microscope. During the collection, sections were placed on slides in chronological order, representing the rostral-to-caudal progression. All analyses presented herein were conducted on the right CB, examined unilaterally.

Multiplex fluorescence labelling

A hydrophobic barrier was drawn around the sections using the ImmEdge® PAP Pen (H-4000; Vector Laboratories) and slides were rinsed in PBS to wash off the OCT. Subsequently, sections were incubated in permeabilization-blocking solution containing 5% normal donkey serum (v/v), 5% normal goat serum (v/v) and 0.3% Triton X-100 (v/v) solution in PBS for 2 hours at room temperature. Sections were rinsed in PBS and incubated in primary antisera solution in 5% normal serum at 4^oC overnight. Primary antibodies were rabbit monoclonal anti-Tyrosine Hydroxylase (TH; 1:500; AB152; Merck) and mouse monoclonal anti-Neurofilament 160kDa (NF) Antibody (1:125; 2H3; DHSB). Sections were rinsed 3x10 min in PBS and incubated in the secondary antisera solution containing secondary antibodies, Lycopersicon esculentum (tomato) lectin conjugated to DyLight®-647 (1:250; DL-1178-1; Vector Laboratories) and DAPI (1:10000; D9542; Sigma-Aldrich) for 2 hours at room temperature. Secondary antibodies used in the study were Donkey anti-mouse IgG AF594 (1:500; A21203; Invitrogen) and Goat anti-rabbit IgG AF488 (1:500; A11008; Invitrogen). Lastly, sections were rinsed 3x10 min in PBS and mounted in

ProLong™ Diamond Antifade Mountant (P36970; Invitrogen) under No. 1.5H precision cover glasses (0107242; Marienfield Superior) and left to set overnight protected from light at room temperature. No primary antibody control (NPC) immunolabelling was performed for each antibody used in the study during pilot optimisation experiments and consistently resulted in an absence of immunofluorescence.

Widefield fluorescence microscopy and image processing

Prepared slides were imaged at the University of Auckland, Biomedical Imaging Research Unit (BIRU) using an Olympus SLIDEVIEW VS200 ASW 4.1 imaging system equipped with Hamamatsu ORCA-Flash4.0 V3 Digital CMOS camera. Multichannel images were acquired using fluorescence filter sets for DAPI (Ex: 365/20 nm; Em: 450/40 nm), FITC (490/20 nm; 525/36 nm), Cy3 (555/25 nm; 590/20 nm) and Cy5 (645/30 nm; 700/50 nm) powered by XYLIS XT720S LED illumination. Images were acquired using an Olympus UPLXAPO 40x /0.95 air objective. Multiple fields of view were tiled together to capture the entire CB area defined manually from an overview image. All images were captured as nine optical sections (z-stack), spaced 0.42 µm apart and centred on a manually set focal point for each section. Exposure was kept constant across all images. Resulting images were exported as 16-bit .TIFF and processed using ImageJ/Fiji software. 

Images were flattened as a maximum intensity projection. Each channel was then deconvolved using the Richardson-Lucy algorithm for 25 iterations using DeconvolutionLab2.²³ Point spread function (PSF) used for deconvolution was generated using the Born and Wolf 3D optical model using PSF Generator.²⁴ In resulting multichannel images CB was selected manually in each image to outline the perimeter of the organ, primarily based on TH immunoreactivity.

Resulting CB outlines are embedded in the images as a selection and overlay.
Overlay can be removed in ImageJ: Image>Overlay>Remove Overlay

Image notation:
date_animal-ID_section-number.tif